I have a pt with exactly the same scenario. I had to look on up-to-date. I don't think there is any study that looked at risk of pulmonary embolism with catheter associated thrombus. I plan to have the patient on 3 months warfarin then stop. The catheter is more on the contralateral side.

I don't think this has any significant risk for PE. That's really not the issue. It's mainly a symptomatic thrombophlebitis. I would have handled this differently:
1. Would NOT remove catheter; no advantage or risk reduction in doing so, only invite more trouble on the left
2. Treat with Eliquis (my preference) or short term SQ Lovenox while adjusting Coumadin (X 3 months); would not need admission
3. Asymptomatic thrombus I might treat if large/occlusive; not sure how you would have found asymptomatic thrombus though
Ted Saad

Thrombus formation around catheter is a natural phenomenon and when one looks for it can be found in many cases.

When there are no symptoms I would not anticoagulate but find an alternate access solution soon.

Sometimes AFTER catheter removal, especially femoral catheters, there will be some wall-adherent thrombus and much discussion can be had weather this is a "DVT".

For management of symptomatic thrombus: if a catheter is needed then we usually leave the catheter in place and start anti-coagulation. - Why? - As thrombus formation is common, patients who develop symptoms (and are less frequent) likely have other pre-disposing factors. Therefore, they have a reasonably high chance to also develop thrombus at the other, new catheter insertion side. This risk may be attenuated by Heparin/Coumadin, but at the cost of additional procedure, use of another access site, initiation central vein stenosis process on the both sides.

Once you commit to anticoagulation we typically continue for at least 1 month after catheter removal AND angiographic resolution of thrombus (if extension into axillary/SCV/BCV). - For IJ thrombus we continue for 1 month after catheter removal and find that this vein typically does not re-canalize if totally occluded (local valves, plenty of collaterals).

I am curious about the use of Xa inhibitors (such as Eliquis) in dialysis patients. Some sources say it is contraindicated in dialysis patients. Can this and the other Xa inhibitors be used in these patients ?

I anticoagulate patients with IJ vein thrombosis secondary to catheter for 3 months duration and I don't remove the catheter. I think it is important to anticoagulate these patients found with acute venous thrombosis secondary to catheter, first to preserve this vein to continue using it for vascular access, second to avoid post thrombotic syndrome (swelling , pain) and third to avoid infections. Timsi et al found more catheter related sepsis in catheters with thrombosis (19%) versus catheters without thrombosis (7%)

I had a.similar case a year ago, he was admitted and anticoagulated initially with heparin and later discharged on coumadin. The thrombus was relatively larger and was 1also confirmed with CTA. 6 months later the clot was completely cleared. The risk of PE may be relatively lower, but he clot size should be carefully looked at.I think it is safe to admit and anticoagulate selected patients

Great to see so many varied opinions and no complaints consensus and barely any large or even medium size studies. Another one to look at for the academic centers for a study. This is something we all see regularly and treated based on who gets to see the patient first. I agree one should not remove the catheter in haste as we may completely loose the site. I have see thrombus around the catheter. It's the nature of placing a catheter in a vein. It's expected as hemostasis is achieved. It's how often we look for it?? And eloquis v/s heparin or lovenox or Coumadin. Again personal preference. We need a position paper.

Thanks for everyone’s comments. Obviously a common scenario which has not been studied.
Even asymptomatic thrombus is frequently detected when other specialties order neck ultrasound or chest CT for a variety of reasons and results in panic.

If catheter is to be left in place, it may be reasonable to anticoagulate and revisit in a month with an echocardiogram. If the thrombus is large, there is a small but definite risk of pulmonary embolism. It is not even clear if it is safe to remove that catheter. I am also cautious about Xa inhibitors in dialysis patients, though these agents are more commonly being used in Europe.

Please see this article (one of many) which discusses the issue- not necessarily fully relevant to nephrology. http://circ.ahajournals.org/content/126/6/768

I do not think we should ignore this. It should be studied systematically and ASDIN can possibly facilitate a study design. This is a large knowledge gap.
Thanks for posting the case and for a nice discussion.
Anil

I generally followed the 2012 Chest guideline: ANTITHROMBOTIC THERAPY AND PREVENTION OF THROMBOSIS, 9TH ED: ACCP GUIDELINES. CHEST 2012; 141(2)(Suppl):7S–47S

9.3.1. In most patients with UEDVT that is associated with a central venous catheter, we suggest that the catheter not be removed if it is functional and there is an ongoing need for the
catheter (Grade 2C) .

9.3.2. In patients with UEDVT that involves the axillary or more proximal veins, we suggest a minimum duration of anticoagulation of 3 months over a shorter period (Grade 2B).

I do agree that this might not be fully relevant to nephrology and further study is required.

Woei Tan

I personally have not anticoagulated these patients.

The primary question here goes back to the risk/ benefits of anticoagulation.

I’ve actually been more concerned about the risks of bleeding in our ESRD/ elderly patients. The incidence of falls/ GI bleeds/ etc. is fairly high for our patients – and most of the time I can’t justify anticoagulation here – especially in asymptomatic patients. We’ve all managed hospitalized ESRD patients with complications of bleeding from anticoagulation.

I have never encountered a symptomatic pulmonary embolism from a thrombus like this. As previously mentioned, thrombus is almost always present with a catheter. I’d be interested to see if others have seen anything symptomatic from an IJ thrombus otherwise.

The question here is if anti-coagulation really benefits the patient, or if we are just treating ourselves with the decision to anticoagulate a potentially benign thrombus….


As a bit of a tangent – I can’t count how many times I’ve stopped anticoagulation orders for ESRD patients who have been vein mapped in the hospital, and have been found to have asymptomatic thrombus at a peripheral upper arm vein (either from peripheral IVs/ old PICC lines/ or previous AV accesses.) Many of the hospitalists have little experience with AV access, and as soon as a Doppler report says the word “thrombus,” the initial reaction has been to treat with anticoagulation. This really does not seem to benefit the patient, and I wonder if others have had the same experience.

Danny Patel

I totally agree that asymptomatic patients should not be anticoagulated. I do think patients with symptomatic DVTs associated with tunneled catheters should have a work up for a hypercoagulable state. I have found symptomatic DVTs associated with tunneled catheters to be fairly uncommon. The work up for a hypercoagulable state to have found some underlying clotting disorders.

Those of you that have imaged patients after removing the catheter, have you found any evidence of vascular thoracic outlet syndrome (similar to Paget-Shcroetter Syndrom). I have found several patients with significant VTOS who developed catheter-associated thrombus after the catheter was placed. Some of these have required first rib release due to there severe VTOS and associated central venous stenosis.

http://aop.sagepub.com/content/48/12/1667.extract
My question is coumadin versus apixaban? If we follow the ACCP guidelines and decide to anti-coagulate for 3 months, and the patient's insurance covered apixaban, What would you choose?

Great case.
First, why does the patient have a CVC at all? Late recognition of ESRD the usual reason; earlier recognition and execution of an effective permanent access plan could have prevented this. Even when patients have no warning of their problem, an early cannulation graft (Acuseal or Flixene) can be usable within 24 hours. Or maybe CAPD as a bridge to a fistula. There is no mention of a permanent access plan.
Second, this is mostly a mechanical problem - a large foreign object in the IJ - and chasing hypercoaguability is a low yield pursuit. I wouldn't bother at this stage of the game.
Third, I agree with Dr. Saad about not moving the catheter. The incidence of DVT/stenosis doubles or quadruples on the left side - and this is where most people want their access, on the left. I agree with anticoagulation.
Fourth, I agree with Dr. Agrawal that PE from the IJ is a low occurence consequence - but maybe not so true of innominant or SVC thrombus. I have seen some very scary "Oh, shit" filling defects in the SVC. Admit for TPA infusion, then anticoagulation.
Get a functioning access in, and get the catheter out ... get the catheter out ... get the catheter out.

Everything said by everyone is of interest but the fact remains there is no good data. Will a clot associated with a catheter propagate or embolize? We don't know. Will anticoagulation while leaving the nidus in place (the tunnelled catheter) do any good? We don't know. Will removing the catheter and anticoagulating (and who knows for how long) restore / maintain patency of that IJ for future use? We don't know. If a patient developed a clot and had therapy that improved things and then gets a catheter placed on that same side at avfuture time, are they at a higher risk for clot with that new line? We don't know. And so on......

There is some investigation into newer anticoagulant agents in CKD and ESRD, so at least we may soon know something about that. We do know that ESRD patients are at higher risk for significant complications from warfarin (and presueably other OAC's but again we don't really know that).

So, this is a great area forvresearch and for guidelines. My own approach is to assess and treat each patient and each event based on that patient's unique status and my own best judgement. FWIW, I onlybtreat these clots if they are symptomatic.

I agree with Marc WEBB about the permanent vascular access, but I prefer a fistula.
There was few place for changing the catheter, with the risk of clotting for the other side central veins!
With a symptomatic patient, I would give coumadine for 3 months

Great discussion. I have pondered this scenario several times over the years, I think I do something different every time. Just some observations.

1. Once I removed a tunneled HD catheter in an asymptomatic patient (at the bedside in the access center) who had successfully started using a fistula. That evening, she had chest pain and went to the ED. She had a CTA revealing a small sub-segmental pulmonary embolism. No symptoms or findings (ie respiratory or cardiac abnormalities) except for the chest pain. I think they gave her coumadin for a few months. This case made me realize that there could be serious complications with something as simple as removal of a tunneled HD catheter. I assume that this patient had thrombus on the tip of her catheter which embolized when the anchoring catheter was pulled out of the venotomy. If the clot is attached to the vessel wall, it might be less likely to migrate as opposed to if it is just adherent to the catheter tip. I now include PE on my informed consent for catheter removal.

In the above case, there was no indication for a radiologic study prior to catheter removal. However, the next case is one where a patient presented to me with a poorly functioning tunneled HD catheter. I performed an angiogram revealing the a large clot around the SVC and catheter tip. Again the catheter is not working. Should I a) remove the RIJ tCVC and put a new one in the LIJ (and risk embolism), b) put a femoral tCVC in place and anticoagulate with coumadin until the clot is gone, then remove the RIJ tCV, c) admit to the hospital for a tPA infusion, d) perform a guidewire exchange of the RIJ tCVC and macerate the thrombus with a balloon? I chickened out by running a Rosen through the catheter ports, to get it working, then started coumadin. I expect the patient back in a few weeks and then will have to make a decision.

Thanks again for a good discussion, these are just two similar cases that i wanted to share. I'll send images of the catheter clot if I can figure how to upload them to this blog.

Cutberto Cortés.
Hi everyone. Good case. I just to add we need to consider the risk or danger of development or worsening VCS when catheter is removed and compromising the sewer system of cephalic circulation.