Bloody LP

This forum made possible through the generous support of SDN members, donors, and sponsors. Thank you.

Bostonredsox

Full Member
10+ Year Member
Joined
Mar 15, 2011
Messages
1,235
Reaction score
120
Is it possible to hit a small segmental branch of some artery coming off the abd aorta near the L5 nerve roots??

Patient - 72 year old admitted a few hours ago in Status Ep. No history if seizures. 10 min seizure at NH, brought to ED, seized again, agonal post ictally, pH 7.1 intubated, lactate near 7. History of HIV supposively taking abacavir. admitted to MICU.

I pick him up a few hours later, while putting a line in him notice he is hypoxic with sats dropping on vent. unhook him and bag him, replace his tubes and leads and sats look better. But hes quivering in his face and lips chattering despite temp of
100.1 Give 4 more of ativan and quivering stops. Dilantin and versed drips going. I call hospitalist and ask for permission to transfer to neuro ICU (we have no neuro coverage this weekend and dont have a single eeg yet). she agrees i call and quickly pt is accepted. They ask for LP and meningeal coverage given HIV status with status.

I do LP. somewhat tough as hes 240 lbs on the vent. Get in...pure red CSF comes back. not 2-3 drops red tinge then clear, dripping blood. Withdraw my needle and get a new needle. re approx landmarks, restick maybe 1 cm away from previous spot....8-10 CC of pure red fluid dripping into tubes. hmmmm maybe a SAH that CT missed? Send to stat Ct....no bleed. Hmmm. Call lab to see what RBC counts were...fluid had clotted in tubes.

Is it possible that my needle was sitting in some small arterial branch and not in the CSF space? I have never seen this before.

Members don't see this ad.
 
You probably hit the peridural venous plexus. Was the patient on any blood thinners? The patient probably needed an MRI with contrast of their L-spine to rule out an epidural hematoma as a complication. It's rare, but it does occur.

If it's bloody CSF, you would think of a SAH or a herpetic encephalitis. If the CSF tubes clotted, then that's pure blood and not a good sign.
 
You probably hit the peridural venous plexus. Was the patient on any blood thinners? The patient probably needed an MRI with contrast of their L-spine to rule out an epidural hematoma as a complication. It's rare, but it does occur.

If it's bloody CSF, you would think of a SAH or a herpetic encephalitis. If the CSF tubes clotted, then that's pure blood and not a good sign.

yes to coumadin but INR was subtheraputic at 1.5

And as for a venous plexus, this was bright red blood. granted hes on the vent with the previous paO2 before adjustment of 225.

I do not have inhouse MRI capabilities at night but I had called back the accepting intensivist about it and asked his thoughts about what could have happened so that he could followup on it. The patient should already be there by now.

And was this a technical error on my part? Is my needle access point to far lateral? did I pass the needle too far? Anything I can do differently to try and prevent it from happening again? and thanks for the input.

And after doing some reading, this may have been better posted in the gas forum as it seems accidental puncture of the interior and exterior venous plexi are most commonly seen with epidural placements. Looking at the diagrams it seems odd to me it doesnt happen more frequently the veins are all over in a tight space.
 
Last edited:
Members don't see this ad :)
It's hard to comment on the technique, since none of us were there or could see what you did. I guess if you were too lateral, in an anticoagulated patient, if you got into muscle you could get frank blood. But if you're that far lateral, you wouldn't hit the interspinous ligament and if you've done a lot of LPs, which I assume you have, it would feel different (less resistance). One thing you can do at the bedside is look for double ring sign. It's usually used to assess for CSF rhinorrhea, and it's not a perfect test, but the idea is similar and you could do it with just a drop or two of fluid on your paper drape.
 
I can't comment on this specific case, but speaking in generalities:

1-When this happens usually you are not intrathecal. The needle could be epidural (epidural venous plexus) or retro spinal. In an anticoagulated patient, you wouldn't needle a large vein to produce pure blood.

http://radiographics.rsna.org/content/24/1/193/F3.expansion.html

2-Even with a SAH you should be able to tell the difference between blood and bloody CSF. Like the above poster said, you just put a drop of it on your sterile drape and see if you get the double ring. I've had to do this before. It never turned out to be CSF, but only blood. That much blood would be seen on CT easily.

3- INR >1.4 is a relative contraindication to LP. You are borderline, and it's a life threatening situation so it's "relative" and you have to make a decision.
 
Last edited:
You can hit the venous plexus and it will look relatively bright compared to blood drawn from a line. Once you've hit a vein, there tends to be at least a small hematoma that collects so you're usually stuck hitting it again if you don't move up at least a level on the second attempt.
 
  • Like
Reactions: 1 user
I don't think this is necessarily a problem with your technique, but not seeing it I can't specifically comment on it. This is a known complication of LP's. If you do enough of them, you will eventually cause an epidural hematoma. It's going to happen. The rate of detecting a SAH not diagnosed by CT is 1 in 700 if I remember correctly. It's been a while since I've read the literature, but I think the rate of epidural hematoma with LP is 1 in 500.

Our malpractice carrier recommends as follows for performing LP's on anticoagulated patients:

ASA, NSAIDs, Aggrenox: safe to perform
Warfarin: wait 5-7 days AND INR < 1.4
LMWH: low dose, wait 12 hours; high dose wait 24 hours
Plavix, Effient: wait 7 days
Ticlid: wait 10 days
Aggrastat, Integrillin: wait 8 hours
ReoPro: wait 24-48 hours
Lepirudin, Angiomax, Argatroban: wait 8-10 hours
Pradaxa: unsafe, do not perform LP
Arixtra: wait 36 hours
Xarelto: wait 20 hours
Danaparoid: unsafe, do not perform LP

If you perform an LP against their advice, you'll probably find yourself without insurance coverage given its their recommendation.

These are the sources they cite (not independently reviewed by me):

Vandermeulen, Erik MD. Best practice and research clinical Anesthesiology. 24 (2010) 121-131
Johnson, Kimberly MD; Sexton, Daniel J MD; UpToDate. Lumbar Puncture: Technique; Indications; Contraindications and Complications in Adults
Layton, Kennith MD; Kallries, David MD; Horlocker, Terese MD. AJNR March, 2006. 27: 468-470
 
So waiting 10 days is not safe with Pradaxa? What about a month? A year? After ten days the concentration should be down one million fold. I don't know about these recommendations. And I'd never heard of a malpractice carrier having practice guidelines before.
 
I don't think this is necessarily a problem with your technique, but not seeing it I can't specifically comment on it. This is a known complication of LP's. If you do enough of them, you will eventually cause an epidural hematoma. It's going to happen. The rate of detecting a SAH not diagnosed by CT is 1 in 700 if I remember correctly. It's been a while since I've read the literature, but I think the rate of epidural hematoma with LP is 1 in 500.

Our malpractice carrier recommends as follows for performing LP's on anticoagulated patients:

ASA, NSAIDs, Aggrenox: safe to perform
Warfarin: wait 5-7 days AND INR < 1.4
LMWH: low dose, wait 12 hours; high dose wait 24 hours
Plavix, Effient: wait 7 days
Ticlid: wait 10 days
Aggrastat, Integrillin: wait 8 hours
ReoPro: wait 24-48 hours
Lepirudin, Angiomax, Argatroban: wait 8-10 hours
Pradaxa: unsafe, do not perform LP
Arixtra: wait 36 hours
Xarelto: wait 20 hours
Danaparoid: unsafe, do not perform LP

If you perform an LP against their advice, you'll probably find yourself without insurance coverage given its their recommendation.

These are the sources they cite (not independently reviewed by me):

Vandermeulen, Erik MD. Best practice and research clinical Anesthesiology. 24 (2010) 121-131
Johnson, Kimberly MD; Sexton, Daniel J MD; UpToDate. Lumbar Puncture: Technique; Indications; Contraindications and Complications in Adults
Layton, Kennith MD; Kallries, David MD; Horlocker, Terese MD. AJNR March, 2006. 27: 468-470

Don't forget about Brillinta (ticagrelor)
 
I would say you hit a small vein or artery. Unlikely to be a significant complication. I've had this happen maybe 5 times. I would move up or down 1 space and try again.
 
I would say you hit a small vein or artery. Unlikely to be a significant complication. I've had this happen maybe 5 times. I would move up or down 1 space and try again.

happened to me once.... incidentally (i think?) also an HIV+ pt.

we have IR at my hospital and pt was being admitted anyway and already got abx... i had them do the repeat. wasn't going to change management from the ED.

no other complication of which i am aware. it happens, as others have said.
 
So waiting 10 days is not safe with Pradaxa? What about a month? A year? After ten days the concentration should be down one million fold. I don't know about these recommendations. And I'd never heard of a malpractice carrier having practice guidelines before.
Their recommendations, not mine.
 
I don't think this is necessarily a problem with your technique, but not seeing it I can't specifically comment on it. This is a known complication of LP's. If you do enough of them, you will eventually cause an epidural hematoma. It's going to happen. The rate of detecting a SAH not diagnosed by CT is 1 in 700 if I remember correctly. It's been a while since I've read the literature, but I think the rate of epidural hematoma with LP is 1 in 500.

Our malpractice carrier recommends as follows for performing LP's on anticoagulated patients:

ASA, NSAIDs, Aggrenox: safe to perform
Warfarin: wait 5-7 days AND INR < 1.4
LMWH: low dose, wait 12 hours; high dose wait 24 hours
Plavix, Effient: wait 7 days
Ticlid: wait 10 days
Aggrastat, Integrillin: wait 8 hours
ReoPro: wait 24-48 hours
Lepirudin, Angiomax, Argatroban: wait 8-10 hours
Pradaxa: unsafe, do not perform LP
Arixtra: wait 36 hours
Xarelto: wait 20 hours
Danaparoid: unsafe, do not perform LP

If you perform an LP against their advice, you'll probably find yourself without insurance coverage given its their recommendation.

These are the sources they cite (not independently reviewed by me):

Vandermeulen, Erik MD. Best practice and research clinical Anesthesiology. 24 (2010) 121-131
Johnson, Kimberly MD; Sexton, Daniel J MD; UpToDate. Lumbar Puncture: Technique; Indications; Contraindications and Complications in Adults
Layton, Kennith MD; Kallries, David MD; Horlocker, Terese MD. AJNR March, 2006. 27: 468-470

These policies are typically adopted from the American Society of Regional Anesthesia recommendations for neuraxial analgesia, ie, epidurals/spinals (see 2010 ASRA Guidelines). Some of the drugs are so new, with new anticoagulants coming out every day it seems, that the policies for some of the drugs are a work in progress.
 
Thanks for all the replies guys. In this case the LP was essentially emergent given his critical state. What you have said makes sense and I guess in hindsight I should have moved up a level after the first stick. I'm quite sure I was not too lateral, he had very prominent spinous processes and i cold palpate quite easily despite his size and non optimal positioning given the vent. But I probably introduced the second needle right into the hematoma from the first.
 
...don't forget Eliquis (another new anticoagulant)
 
Top