Fresh SAH or bloody tap?
Started by tkim
Get help with your application
Use all the free resources available to you from SDN: articles, guides, expert advising, forums discussions, and school research.
I'm going with traumatic tap. You would expect xanthochromia in SAH (though not 100%, especially right away.) Also would expect higher WBC:RBC in SAH. Also, RBCs drop from tube 1-->4 in bloody tap, not SAH.
Advertisement - Members don't see this ad
D
deleted109597
I would argue that there needs to be more story to this. How many attempts? What did it look like to the person performing?
There isn't good actual data out there to tell you one way or the other based on how the RBCs trend down. You can also have a traumatic SAH tap, they aren't out of the realm.
Lack of xanthochromia is promising unless they said sx started 1 hour PTA. Then you're just ****ed.
There isn't good actual data out there to tell you one way or the other based on how the RBCs trend down. You can also have a traumatic SAH tap, they aren't out of the realm.
Lack of xanthochromia is promising unless they said sx started 1 hour PTA. Then you're just ****ed.
You only see xanthochromia after at least 12 hours.
Where I work the answer is simple.
I tuck my tail between my legs and call neuro. Tell em im worried about SAH and this tap was a mess. the use of LP for SAH with much if any blood in the CSF is just inconclusive. Even the number of RBCs is pretty unclear. Is it less than 5? so if you have 6 rbcs is that a SAH.
Anyways Call neuro, admit, they usually get MRI/MRA or CTA and perhaps a CT guided LP.
I tuck my tail between my legs and call neuro. Tell em im worried about SAH and this tap was a mess. the use of LP for SAH with much if any blood in the CSF is just inconclusive. Even the number of RBCs is pretty unclear. Is it less than 5? so if you have 6 rbcs is that a SAH.
Anyways Call neuro, admit, they usually get MRI/MRA or CTA and perhaps a CT guided LP.
With that CSF analysis, I'd be extremely worried about just chalking it up to a traumatic tap.
If I'm doing an LP for HA, then I have at least a medium pretest probability for SAH (assuming there are no neuro abnormalities--which should show blood on scan). It's kind of like getting a dimer where the result is at the cutoff, but the history is someone with OCPs, smoking, acute onset, tachy, chest pain, etc. At that point, I think you need to escalate the workup.
Again, assuming there's no neuro abnormalities, if there's an initial bleed, it should tamponade quickly given that you're not seeing any clinical deterioration. Generally, as long as they're stable, you've got time to consult.
I don't know the evidence, but I think this is the ideal situation for CTA or MRA. Obviously, it's not initially warranted in the patient being worked up for HA since 3-5% of general population walks around with a cerebral aneurysm and you would only diagnose existence rather than rupture. However, with what I presume was a convincing story and that tap, it would be worth it, along with a call to neuro or neurosurg. Given the devastation of a possible subsequent bleed, they may want to go right to invasive angio.
Overall, if I had a convincing story, blood on a tap, and anything on imaging or conventional angio, I'd want it coiled, even if I could never be sure that exact aneurysm bled in the first place.
Of course, with all that said, my last SAH patient had a classic story, no blood on CT and refused LP at an OSH. Came back 2 weeks later with repeat HA but normal exam. They did LP first for whatever reason--looked like traumatic tap, no xanthrochromia. On scan that day after the LP, she had blood all over the place. On initial invasive angio here, neurosurg couldn't find an aneurysm and there was nothing abnormal on MRA either. Only after doing a repeat invasive angio one week later did they find a 2 mm superior cerebellar aneursm . . .
If I'm doing an LP for HA, then I have at least a medium pretest probability for SAH (assuming there are no neuro abnormalities--which should show blood on scan). It's kind of like getting a dimer where the result is at the cutoff, but the history is someone with OCPs, smoking, acute onset, tachy, chest pain, etc. At that point, I think you need to escalate the workup.
Again, assuming there's no neuro abnormalities, if there's an initial bleed, it should tamponade quickly given that you're not seeing any clinical deterioration. Generally, as long as they're stable, you've got time to consult.
I don't know the evidence, but I think this is the ideal situation for CTA or MRA. Obviously, it's not initially warranted in the patient being worked up for HA since 3-5% of general population walks around with a cerebral aneurysm and you would only diagnose existence rather than rupture. However, with what I presume was a convincing story and that tap, it would be worth it, along with a call to neuro or neurosurg. Given the devastation of a possible subsequent bleed, they may want to go right to invasive angio.
Overall, if I had a convincing story, blood on a tap, and anything on imaging or conventional angio, I'd want it coiled, even if I could never be sure that exact aneurysm bled in the first place.
Of course, with all that said, my last SAH patient had a classic story, no blood on CT and refused LP at an OSH. Came back 2 weeks later with repeat HA but normal exam. They did LP first for whatever reason--looked like traumatic tap, no xanthrochromia. On scan that day after the LP, she had blood all over the place. On initial invasive angio here, neurosurg couldn't find an aneurysm and there was nothing abnormal on MRA either. Only after doing a repeat invasive angio one week later did they find a 2 mm superior cerebellar aneursm . . .
The story:
37 y/o F driving car when sudden onset of 10/10 bilat frontal ha within minutes became the "worst of my life - and I never get headaches" without prompting on my side, had to pull over and call ambulance. Over time, HA began to radiate to occiput, with mild neck stiffness. No focal deficits. No other complaints. No migraine hx. No meds. Not crazy. Seen about 7 hours after onset HA.
Neg CT. LP attempt x 3, initial L3-4, then L4-5 buried to the hub, then again at L3-4 with wine-colored, cloudy fluid without any appreciable clearing with subsequent tubes. LP performed seated, so no opening pressure. CSF protein 195.
Shipped to tertiary hosp. No word back. Gonna follow up in a day or so.
I've done tramatic taps, but all of them have cleared by tube 4. This one was different. Bloody all the way.
37 y/o F driving car when sudden onset of 10/10 bilat frontal ha within minutes became the "worst of my life - and I never get headaches" without prompting on my side, had to pull over and call ambulance. Over time, HA began to radiate to occiput, with mild neck stiffness. No focal deficits. No other complaints. No migraine hx. No meds. Not crazy. Seen about 7 hours after onset HA.
Neg CT. LP attempt x 3, initial L3-4, then L4-5 buried to the hub, then again at L3-4 with wine-colored, cloudy fluid without any appreciable clearing with subsequent tubes. LP performed seated, so no opening pressure. CSF protein 195.
Shipped to tertiary hosp. No word back. Gonna follow up in a day or so.
I've done tramatic taps, but all of them have cleared by tube 4. This one was different. Bloody all the way.
Not sure how you could've done anything different. tons of RBC's with very concerning story = SAH until proven otherwise.
Not exactly. The sensitivity of xanthochromia goes up with time, and doesn't approach 100% until 12h. At 6 hours, it's about 60%. So the absence of xanthochromia should only be used to exclude SAH LP after 12h, but you can see xanthochromia in CSF before 12h.
I don't know. I would call nsurg and do mri.
And if that were negative for bleed or aneurysm, what's your dispo? Would you be comfortable discharging a patient with a bloody tap and no source? This is a community shop with no nsurg avail. I'm not sure I would be comfortable discharging that patient without a neurosurgeon seeing the patient.
Last edited:
Plus, many labs don't assess xanthochromia correctly. They technically need to use a spectrophotometer in order to call it colorless. Plenty of labs actually just hold the tube up to a piece of paper.
Very true.
Again, I'm even more freaked out now after having my last SAH come in with blood on her scan and no aneurysm on invasive angio or MRI--only on repeat angio a week later . . .
Advertisement - Members don't see this ad
Based on the tap information alone, I would have asked you for a history and physical. With the information supplied, I would certainly be concerned for SAH, reversible cerebral vasconstriction, venous thrombosis, or primary thunderclap headache. A young person with no medical history and a sudden-onset severe HA is concerning (an understatement).
Obviously a high concern on the list is SAH (from everyone's perspective, whether neurologist or EM).
With a negative CTH, I'd proceed to CTA/V and/or MRI/A/V. If positive, you have your answer, if negative...I would want to transfer to a place with NSG for an observation admission. I don't see how anyone could fault you for transferring with a concerning story like this. And primary thunderclap headache is a diagnosis of exclusion.
Obviously a high concern on the list is SAH (from everyone's perspective, whether neurologist or EM).
With a negative CTH, I'd proceed to CTA/V and/or MRI/A/V. If positive, you have your answer, if negative...I would want to transfer to a place with NSG for an observation admission. I don't see how anyone could fault you for transferring with a concerning story like this. And primary thunderclap headache is a diagnosis of exclusion.
"A 25% reduction in RBC concentration between the first and fourth tubes of CSF in patients with suspected SAH, but negative head CT findings, occurs even in cases of ruptured aneurysms. Formal evaluation for aneurysm is still necessary in this scenario. "
-AJNR 2005 26: 820-824
-AJNR 2005 26: 820-824
I've been taught that the RBC count has to go down to almost 0 to confidently exclude bleeding. Also, I read somewhere that the majority of hospitals do not detect xanthrochromia in the appropriate manner using spectro.
I would've called neurosurgery. At my hospital, usual practice would be CT angio and discharge with phone c/s if negative. I've had pts with positive taps sent to our ER from other smaller hospitals. Neurosurgeon usually requests CT angio and discharges if neg...usually without seeing pt.
Having said that, I can recall 2 cases of SAH with neg CTA. One even had a neg tap. I've never found or known anyone to dx subarach bleed with an LP. But the general consensus from what I know is we still have to do the tap.
I would've called neurosurgery. At my hospital, usual practice would be CT angio and discharge with phone c/s if negative. I've had pts with positive taps sent to our ER from other smaller hospitals. Neurosurgeon usually requests CT angio and discharges if neg...usually without seeing pt.
Having said that, I can recall 2 cases of SAH with neg CTA. One even had a neg tap. I've never found or known anyone to dx subarach bleed with an LP. But the general consensus from what I know is we still have to do the tap.
Do any of you use CTA instead of LP to rule out SAH? Of course it detects the aneurysm, but you never know if it's bleeding until you LP. I've gotten into the practice of performing a CTA. If it's negative, I stop there. If there's an aneurysm, then I perform an LP.
I always have a discussion with the patient about this, and most (if not all) choose the CTA over the LP.
On one case that I thought was a definite SAH that I would've missed by LP, ended up being a thrombosis.
I always have a discussion with the patient about this, and most (if not all) choose the CTA over the LP.
On one case that I thought was a definite SAH that I would've missed by LP, ended up being a thrombosis.
So in your experience patients prefer a passive imaging test instead of a giant needle shoved into their spine? 🙂
Do you have literature to support your practice? I have seen one SAH on LP that I didn't see on CT. It's a condition with a low prevalence, so not testing for it means you'll be OK for quite a while. Besides, who is to say the sentinel bleed that you missed was really a sentinel bleed since no gold standard test was performed?
I ordered an MRA on a patient who didn't want an LP and the radiologist said "a negative MRA does not rule out the presence of acute SAH or an aneurysm and further clinical testing should be done if this is a concern". Youch.
Curious as to why you proceed to an LP? Does it add anything to the clinical management? Why not just get NSGY or NIR to take care of the aneurysm?
"In ED patients complaining of acute-onset headache without significant SAH risk factors"
OK, so in a low-risk population, it works great. The post I was referring to seemed to suggest his standard practice was CT/CTA and then stop, regardless of the patient.
It's not *MY* practice - I still CT/LP.
Late to this one and didn't read all the posts, but from the original info, I'd say... SAH unless proven otherwise, with no other information. That RBC count on the 4th tube isn't dropping enough to have any sort of confidence in calling it a traumatic tap. You just can't do it. Gotta assume SAH unless someone more specialized than I calls it otherwise. My take. Can't rely on the xantho either.
D
deleted109597
Because 2-5% of people have aneurysms that aren't bleeding, and NSGY isn't a benign procedure.
I have to agree with Southerndoc in that at negative CT/CTA has a low enough risk for LP that you can get away without doing it. A negative CT/Positive CTA needs the tap, and you can likely reduce the number of LPs by starting out with CTA.
Remember, an LP is easy on small, skinny people. Since none of those actually come to my ED, an LP is a long drawn out procedure of me hubbing the longest 20 gauge I have x3-4 attempts, then waiting for their INR so that IR will do it. A CTA shortens that LOS prodigiously.
What about ruling out the non-aneurysmal SAH?
D
deleted109597
Less common than silent MIs (if you put traumatic SAHs in another category), which still stand at 2%.
UTD pegs it at 15-20%
Missing 15-20% of bleeds guarantees a lawsuit in your future.
D
deleted109597
You wouldn't miss 15-20% (which is likely skewed high by SAH without a found aneurysm). You would miss 15-20% of the 5-10% you don't pick up with the initial CT. So you're talking about 0.75-2% miss rate, which is likely what we are missing anyway. I still offer them the LP, but feel better about it when they inevitably refuse.
Now, maybe your shop is different, but I'm not doing an LP every day, and I certainly have plenty of people with headaches. It's not like I'm sending people home to die every day.
Advertisement - Members don't see this ad
I average one LP/week though most are for r/o meningitis. A 2% miss rate is high. But that's beside the point. All it takes is one to ruin your day.
D
deleted109597
We as a specialty miss 2% of MIs.
And every time we do, our butts pucker and we wonder if we should call our carrier.
D
deleted109597
I'm not saying we shouldn't try, but at the same time, our positive chest pain admission rate (for UA/ACS, not just for CAD) is well below 50%. Some things aren't cost effective or time effective, and will become less so in the future as the population ages.
Yes, I offer LP to everyone with negative CT/negative CTA, but the discussion is different than the people who get negative CT/positive CTA. And because the you can tell them the percentages, you have them make their own informed decision. How many people tell their patients that if they have SAH and refuse LP, the second bleed is fatal more often than not? I tell people that, but I still have people refuse. And I get about 1 LP a month, because the vast majority of the headaches that come in are the kind that ask for demerol/phenergan IV push. Or stadol, apparently people used that a lot here before me. I, on the other hand, get to have the nurses ask me what droperidol is every time I order it.
Yes, I offer LP to everyone with negative CT/negative CTA, but the discussion is different than the people who get negative CT/positive CTA. And because the you can tell them the percentages, you have them make their own informed decision. How many people tell their patients that if they have SAH and refuse LP, the second bleed is fatal more often than not? I tell people that, but I still have people refuse. And I get about 1 LP a month, because the vast majority of the headaches that come in are the kind that ask for demerol/phenergan IV push. Or stadol, apparently people used that a lot here before me. I, on the other hand, get to have the nurses ask me what droperidol is every time I order it.
Because if the LP is negative, then its not an acute bleed and can be managed on a less urgent basis (I've sent them home with follow-up with neurosurgery the following day).
Gotcha. Thanks!
Probable traumatic tap. Would have done CTA or MRA.
I actually do a lot more LPs than i did even a few years ago, and definitely more now than I did as a resident. I'm a lot faster at LPs and better at them now (now 5+ years out of residency). I'm not willing to miss a SAH.
It just takes one case to make the rest of your career $hitty.
Q
I actually do a lot more LPs than i did even a few years ago, and definitely more now than I did as a resident. I'm a lot faster at LPs and better at them now (now 5+ years out of residency). I'm not willing to miss a SAH.
It just takes one case to make the rest of your career $hitty.
Q
Now that I've returned to this thread, I am beginning to side with the possibility of a SAH.
