Low pretest probability SAH and a normal CTA Brain...

[pinipig523]

So, I was thinking - you have a guy with not the worst headache of his life, slow onset, but on the occipital aspect of his head. No hx of sah/ich/aneurysm/avm or fhx of sah.

Now, his headache is 7/10.

I did not buy the SAH story on this one.

I did, however, do a CTA Brain to look at his vessels and to see if he had an aneurysm. If he had an aneurysm - I was going to LP him. If there was no aneurysm - then nothing.

Does anyone do this?

---

Members don't see this ad.

 

[Rendar5]

There is a negative potential to this whole CTA business: If you find an aneurysm and the LP is negative, that pt is still gonna end up referred to neurosurg, if not by you then by their PMD. They may decide this is then a symptomatic aneurysm since it was found during a headache workup, and perform potentially unnecessary coiling, stapling, or other procedures.

 

[Arcan57]

There is a negative potential to this whole CTA business: If you find an aneurysm and the LP is negative, that pt is still gonna end up referred to neurosurg, if not by you then by their PMD. They may decide this is then a symptomatic aneurysm since it was found during a headache workup, and perform potentially unnecessary coiling, stapling, or other procedures.

I'm not sure I buy the "neurosurgery will butcher the patient" argument as a completely convincing reason for not doing a CTA on patients. I think a better reason is that you have no suspicion that they have a leaking aneurysm as the source of their headache. The only time I use CTA Brain scans is when I think it's likely they have a SAH and I can't convince them to do the LP.

 

[coachB]

What if it was a non-aneurysmal SAH? I've also seen "negative" CTAs where aneurysms were initially missed.

 

[southerndoc]

There is a negative potential to this whole CTA business: If you find an aneurysm and the LP is negative, that pt is still gonna end up referred to neurosurg, if not by you then by their PMD. They may decide this is then a symptomatic aneurysm since it was found during a headache workup, and perform potentially unnecessary coiling, stapling, or other procedures.

Even if the aneurysm isn't leaking it should still be treated. Doesn't need to be treated emergently, but it needs to be treated: either monitored for growth or coiled.

That's like saying a AAA being detected as an incidental finding shouldn't be treated.

There is evidence that coiling prevents neurologic catastrophe. It might not have been leaking when you detected it, but in ten years time you can't say that the patient isn't at high risk of rupture.

 

[docB]

There is evidence that coiling prevents neurologic catastrophe. It might not have been leaking when you detected it, but in ten years time you can't say that the patient isn't at high risk of rupture.

That's an interesting statement. I work up a headache with concern for SAH by CT then if negative LP then if negative d/c. In that scenario we don't catch aneurysms that aren't leaking. Is there a trend toward doing CTA? I'd swear the last article I read on this (which was part of the LLSA a few years back) said CT/LP was still the way to go.

I'm asking this as a legitimate question to see if I'm doing it wrong . The farther I get from residency the more I worry about being a dinosaur.

 

[pinipig523]

I think for concerning stories - CT/LP is the way to go.

My thinking was that in someone with a low pretest probability of a SAH but is 3 days out (meaning that your sensitivity on a non-contrast CTH would be around 80%), a CTA of the brain would be a great way of really lowering your post test probability even more. If the CTA shows no aneurysms, then in a low risk patient who is not concerning 3 days out, you can get away without an LP.

I was ready to bite the bullet that if I saw an aneurysm, then we do an LP regardless of how low I thought his pretest probability was.

 

[pinipig523]

That's an interesting statement. I work up a headache with concern for SAH by CT then if negative LP then if negative d/c. In that scenario we don't catch aneurysms that aren't leaking. Is there a trend toward doing CTA? I'd swear the last article I read on this (which was part of the LLSA a few years back) said CT/LP was still the way to go.

I'm asking this as a legitimate question to see if I'm doing it wrong . The farther I get from residency the more I worry about being a dinosaur.

Concerning story for SAH - you need to do CT/LP.

Sensitivity of CT non contrast is inversely proportional to time since onset of headache/bleed. 96% sensitivity within 12 hours, 80% sensitivity at 72 hours I think.

The CTA Brain only shows you a roadmap of the vessels, the sensitivity for bleed is still the same as a non contrast CTH. You will still miss bleeds.

 

[southerndoc]

That's an interesting statement. I work up a headache with concern for SAH by CT then if negative LP then if negative d/c. In that scenario we don't catch aneurysms that aren't leaking. Is there a trend toward doing CTA? I'd swear the last article I read on this (which was part of the LLSA a few years back) said CT/LP was still the way to go.

I'm asking this as a legitimate question to see if I'm doing it wrong . The farther I get from residency the more I worry about being a dinosaur.

No, your concern is for a sentinel bleed. CT/LP and if negative, they can follow-up for an outpatient MRI/MRA where it can be arranged.

CTA can be used to r/o aneurysms and thus most sentinel bleeds. Aneurysms <2 mm won't show up, but often active extravasation of contrast will show if it's leaking. Some physicians utilize CTA's instead of LP's for various reasons.

 

[Apollyon]

What if it was a non-aneurysmal SAH? I've also seen "negative" CTAs where aneurysms were initially missed.

That's like a PE with an INR of 4. These things happen, but what are we supposed to do? Your first point is subarachnoid blood coming from where? Somewhere bleeding without a visible structural weakness? Then, I guess it is to the gold standard of angio, or maybe open - I am not an neurosurgeon or neurorads or IR.

And, your second point - do you allude to actually missed, like, it was there, someone looked right at it, and didn't see it, or the slices just didn't catch it - that it slipped through the cracks, and were seen on a subsequent scan?

 

[neusu]

The utility of CTA is really dependent on how good your neuroradiologists are at reading CTAs. MRI/A with FLAIR and a gradient echo/flash are more sensitive but obviously take more time to obtain. The gold standard is still a DSA. The guys here generally noncon CT first, then LP. If the LP is negative the patient goes home. We only get called if the LP is "positive," but from my experience only 1/10 of those actually has an aneurysm and the other 9 have a traumatic tap or some other vasculopathy that goes to neurology (CADASIL etc).

 

[Rendar5]

Even if the aneurysm isn't leaking it should still be treated. Doesn't need to be treated emergently, but it needs to be treated: either monitored for growth or coiled.

That's like saying a AAA being detected as an incidental finding shouldn't be treated.

There is evidence that coiling prevents neurologic catastrophe. It might not have been leaking when you detected it, but in ten years time you can't say that the patient isn't at high risk of rupture.

I dunno if I'd compare it to a AAA honestly as much as I'd compare it to incidentilomas and lung nodules that we pick up on CT's and CXR's. Obviously a sizeable cerebral aneurysm needs treatment as does a symptomatic one. But when 5% of the population supposedly has incidental head aneurysms, I wouldn't go searching for it on someone with a story that doesn't put SAH on my differential. I would go searching for it if my LP was refused, and I'd let the neurologist search for it as an outpatient if I wasn't thinking SAH.

As for the question of non-aneursymal bleeds, the last I heard was that these types of atraumatic bleeds do well without treatment or intervention, but I don't know the source data. Only heard a few people talk about it on EMRAP and emcrit, iirc. Not that I'd recommend taking a chance on these. Just recent commentary that if you miss these small percentage of SAH's, it wouldn't necessarily lead to harm

 

[FDNewbie]

That's an interesting statement. I work up a headache with concern for SAH by CT then if negative LP then if negative d/c. In that scenario we don't catch aneurysms that aren't leaking. Is there a trend toward doing CTA? I'd swear the last article I read on this (which was part of the LLSA a few years back) said CT/LP was still the way to go.

I'm asking this as a legitimate question to see if I'm doing it wrong

. The farther I get from residency the more I worry about being a dinosaur.

docB, you bring up a good point.

This has been a topic of particular interest to me, so I've tried to stay somewhat on top of the latest developments in it (LP vs CTA in *diagnosing* a suspected SAH - NOT its use in mapping for the neurosurgeon). If you look at the articles from only a few years ago ('05 - '07), the consensus was CTA just isn't good enough (yet). But with the advent and proliferation of smaller cut CT scanners, I think the trend is moving toward CTA being an acceptable alternative. Can you say it's accepted as the standard of care? Definitely not. But I think it's only a matter of time when you'll have both options available to you.

The article I like to reference when this discussion is brought up is this one: http://www.ncbi.nlm.nih.gov/pubmed/20370785.

Can computed tomography angiography of the brain replace lumbar puncture in the evaluation of acute-onset headache after a negative noncontrast cranial computed tomography scan?
Done at the Univ. of Buffalo

CONCLUSIONS: CT followed by CTA can exclude SAH with a greater than 99% posttest probability. In ED patients complaining of acute-onset headache without significant SAH risk factors, CT/CTA may offer a less invasive and more specific diagnostic paradigm. If one chooses to offer LP after CT/CTA, informed consent for LP should put the pretest risk of a missed aneurysmal SAH at less than 1%.

And if that isn't enough to make ya say hmmm, here's some food for thought. For the diehard LP fans, kindly keep in mind that:

- LP may be negative less than 2 hours after the bleed. And that's typically well before time of eval & CT for most patients I'M concerned about for an SAH (from time of onset to presentation to the ED via EMS is typically in the 30 minute range, and those are patients I'm getting a Medical Command call to notify me of a possible stroke...so they bypass the ED and go straight to the CT scanner for an emergent dry CT). And if I'm still THAT concerned, I'm doing an emergent LP - within 30 min for sure. So everything's done well within 1 hour - which is in the "blind spot" for definitively diagnosing a SAH via LP.

- LP is most sensitive at 12 hours after symptom onset. That's *definitely* not how long we wait to perform our LPs.

Just my $0.02.

 

[deleted109597]

...(from time of onset to presentation to the ED via EMS is typically in the 30 minute range, and those are patients I'm getting a Medical Command call to notify me of a possible stroke...so they bypass the ED and go straight to the CT scanner for an emergent dry CT). And if I'm still THAT concerned, I'm doing an emergent LP - within 30 min for sure. So everything's done well within 1 hour - which is in the "blind spot" for definitively diagnosing a SAH via LP.

1. Font tags, really?
2. Why is medical command calling a code stroke on "the worst headache of my life", but no neuro findings (photophobia isn't in and of itself a neuro finding). There's no data that picking up hemorrhagic CVAs before 3(4.5)hrs helps. They shouldn't go to the scanner before the history (and some shouldn't at all).
3. CTA is for people who don't want the LP, be it the patient or the doc. I try not to talk myself out of too many LPs. Besides, it is the ultimate in negative reinforcement for narcotic seekers.
4. If you're THAT concerned, just wait until 2 hours. There's no harm. It also gives time for your meds to kick in and the patient is all sleepy, at which point the LP is easier.

 

[deleted109597]

Although...
http://www.madisonrecord.com/news/246245-spinal-taps-were-a-mistake-med-mal-complaint-states
Not that anything in this article makes sense. Treating HTN with Plavix?

 

[FDNewbie]

1. Font tags, really?

It bother you THAT much? I figure we have bigger things to make an issue of

2. Why is medical command calling a code stroke on "the worst headache of my life", but no neuro findings (photophobia isn't in and of itself a neuro finding).

Because they see "stroke" as a general term for the majority of neurologic changes OR complaints that may warrant emergent imaging, be it ischemic or hemorrhagic.

There's no data that picking up hemorrhagic CVAs before 3(4.5)hrs helps.

Kindly elaborate on this.

They shouldn't go to the scanner before the history (and some shouldn't at all).

That's in the hands of the ED Administration. That's the protocol, and that's what happens. It also expedites picking up the bleed, which looks a lot better, regardless of whether it makes a significance or not. When you have a complete H&P, vitals, labs, and consequently a 2 hour delay until the CT is obtained and demonstrates a hemorrhagic stroke, wanna know what the first question outta the patient//patient's family mouth is gonna be? "Why'd you wait so long?" And I'm sure that opens the door to suits for failing to promptly diagnose a potentially life-threatening condition (be they with or without merit).

3. CTA is for people who don't want the LP, be it the patient or the doc. I try not to talk myself out of too many LPs. Besides, it is the ultimate in negative reinforcement for narcotic seekers.

I agree with you on on the 2nd and 3rd points. As for whether a CTA is only for those who don't want the LP, again, I think we should let the evidence speak for itself.

4. If you're THAT concerned, just wait until 2 hours. There's no harm. It also gives time for your meds to kick in and the patient is all sleepy, at which point the LP is easier.

Fair enough, but I'm saying that I bet a fair number - if not the majority - of LPs done for sudden, severe, 10/10 HAs in the ED are done just under this 2 hour window. So if we're gonna chant about how LP is the gold standard, we should at least perform it with the correct methodology (after 2 hrs).

Although...
http://www.madisonrecord.com/news/246245-spinal-taps-were-a-mistake-med-mal-complaint-states

Not that anything in this article makes sense. Treating HTN with Plavix?

Of course not. Unless I missed anatomy class, and the ED docs mistook T5 for L5, she probably suffered zero long-term affects, and only some minor transient effects from a hematoma (which is one of the risks we specifically discuss when consenting for the procedure...).

Ironically, THAT'S probably an excellent patient to do the CTA in, instead of the LP - patients who are anticoagulated who can't be readily reversed, and/or IR isn't available to do it under direct fluoroscopy.

 

[deleted109597]

It bother you THAT much? I figure we have bigger things to make an issue of

Just makes my eyes go all buggy.

Because they see "stroke" as a general term for the majority of neurologic changes OR complaints that may warrant emergent imaging, be it ischemic or hemorrhagic.

Kindly elaborate on this.

Similar to how "trouble breathing" means albuterol and lasix? I mean, let's cover everything with blanket terms. The elaboration is that catching the bleed early doesn't change outcomes. Treatment of the HTN does. How many head bleeds spend all night in the ICU and get operated on the next day (or never)? And you should be treating the HTN (and pain) before the scan to begin with.

It also expedites picking up the bleed, which looks a lot better, regardless of whether it makes a significance or not. When you have a complete H&P, vitals, labs, and consequently a 2 hour delay until the CT is obtained and demonstrates a hemorrhagic stroke, wanna know what the first question outta the patient//patient's family mouth is gonna be? "Why'd you wait so long?" And I'm sure that opens the door to suits for failing to promptly diagnose a potentially life-threatening condition (be they with or without merit).

People sue for lots of reasons. Rushing people to the scanner (and possibly bumping others out of line) isn't going to stop that

I agree with you on on the 2nd and 3rd points. As for whether a CTA is only for those who don't want the LP, again, I think we should let the evidence speak for itself.

Find me some that doesn't say that 5% of the population doesn't have an identifiable aneurysm (that isn't bleeding), and that we don't cause harm to people by putting them in a "CT every 6 months for life" group, then I'll buy it. CTA does help show where the bleed is, after you've identified a bleed. That's something the neurosurgeon cares about. Not something that matters to us. However, I'm more likely to let concede with a patient about not getting an LP if they agree to a CTA. Similarly, there are a lot of docs who want to do CTAs so they don't have to be doing LPs.

Fair enough, but I'm saying that I bet a fair number - if not the majority - of LPs done for sudden, severe, 10/10 HAs in the ED are done just under this 2 hour window. So if we're gonna chant about how LP is the gold standard, we should at least perform it with the correct methodology (after 2 hrs).

I've never gotten an LP in someone less than 2 hours after symptom onset. Hell, they're usually still in the lobby at that point. I would bet more of them fall in the after 12 hour window than under 2 hours.

Ironically, THAT'S probably an excellent patient to do the CTA in, instead of the LP - patients who are anticoagulated who can't be readily reversed, and/or IR isn't available to do it under direct fluoroscopy.

There's also no data for not doing an LP in those patients. Sure, there is a "consensus" that isn't driven by science, but no hard facts. Of course, knowing how reticent everyone is for doing anything with a patient on Plavix, I'm sure they probably didn't even know she was on it (or she said something inane like "something for hypertension")

 

[Blake300m]

I think upwards of 85% of bleeds are aneurysmal, the other 15% (probably less based on the most recent case series) never find a source for the bleed.
Of the bleeds w/o source (or non-aneurysmal), apparently mortality estimated to be 0-15%, with a recent case series (1100 pts, 97 with non-aneurysmal bleeds) having only 4% mortality (Pyysalo LM et al. Long term outcome after subarachnoid hemorrhage of unknown etiology. JNNP 2011) - Finnish paper.

So these people do pretty well whether we diagnose them or not, and not sure how much we can offer them even if we do diagnose a non-aneurysmal bleed other than bp control and a very expensive w/u.

I think once we have a negative CT head, we really need to have the discussion with the pt (if they're capable of following along) and explain risks/benefits. In the OP case, explain to the guy that IF he had a bleed, the CT would have picked it up 70-80% of the time 3 days out; BUT I don't think you have a bleed to begin with. I try to give them some sort of rough percentage, explaining that it's definitely an estimation (so in this guy's case, maybe pre-test prob is 3%, after neg CT it goes to about 1% assuming sn of about 66%). Then explain the options - can do an LP or can do a CTA (with risks of dye and radiation) although 2-8% of population has aneurysms so if we find an aneurysm then we definitely have to do LP to see if it's bleeding; OR can do nothing with the understanding that we haven't 100% ruled out SAH. I've found most people to be very receptive to this, and I think you are covered if you document the conversation (at least in Texas I think I'd be).

I do think there is harm in finding aneurysms on CTA that were never associated with the pt's HA to begin with. ISUIA (International Study of Unruptured Intracranial Aneurysms) is best data we have to work with - Among the patients without prior SAH with posterior communicating, vertebrobasilar/posterior cerebral, and basilar tip UIAs &#8805;25 mm in diameter, the risk of rupture was &#8776;45% at 7.5 years; 10- to 24-mm UIAs and <10-mm UIAs in the same locations carried rupture risks of &#8776;15% and &#8776;2% over 7.5 years, respectively. In all other locations, the rupture risks at 7.5 years for &#8805;25-mm, 10- to 24-mm, and <10-mm UIAs were &#8776;8%, &#8776;3%, and &#8776;0%, respectively.
So aneurysm <10 mm in high risk location, 2% rupture rate over 7.5 yrs; low risk location, roughly 0% over 7.5 yrs

Which would argue who cares if we miss a small aneurysm on CTA, as outcome is good anyway - although the rupture rate of aneurysms in pts with a previous SAH is 10 times higher, so for size <10 mm it goes from 1/2000 per yr to 1/200 per yr (still pretty damn good)

With regards to another poster, I rarely see a pt I'm concerned for SAH come in within 2 hrs, much less 6 hrs - if I had to put a number on it, I'd say >95% come in after 2 hours, and that's being conservative. so the early neg LP while something to consider rarely is actually an issue (plus the CT should be so damn good picking it up at that point that it's a wash, and if the CT is neg I'm comfortable waiting a couple hrs to do their LP)

 

[dotcb]

See the attached ACEP clinical policy on SAH (Level B evidence). Unless you'd like to have it used against you, CT/LP is the way to go. At least it's worth knowing that the CTA approach is not endorsed by our professional organization. In my mind, whether there is an aneurysm or not on CTA does not give you information as to whether it is leaking or the cause of the headache; therefore it does not contribute to the workup in a helpful way.

 

[pinipig523]

See the attached ACEP clinical policy on SAH (Level B evidence). Unless you'd like to have it used against you, CT/LP is the way to go. At least it's worth knowing that the CTA approach is not endorsed by our professional organization. In my mind, whether there is an aneurysm or not on CTA does not give you information as to whether it is leaking or the cause of the headache; therefore it does not contribute to the workup in a helpful way.

ACEP recommends that CT/LP be done to those you are suspecting to have SAH per hx of sudden onset, severe headache.

A normal CTA Brain in a patient w/ low risk, non-sudden onset, not worst headache of his life gives you more ammunition that there is nothing wrong in his head w/ regards to a bleed.

On the other hand, a normal CTA Brain in a patient w/ sudden onset, worst headache ever whom you have a high suspicion for a SAH still needs an LP.

That's my 2 cents, I could be wrong.

 

[pinipig523]

- LP may be negative less than 2 hours after the bleed. And that's typically well before time of eval & CT for most patients I'M concerned about for an SAH (from time of onset to presentation to the ED via EMS is typically in the 30 minute range, and those are patients I'm getting a Medical Command call to notify me of a possible stroke...so they bypass the ED and go straight to the CT scanner for an emergent dry CT). And if I'm still THAT concerned, I'm doing an emergent LP - within 30 min for sure. So everything's done well within 1 hour - which is in the "blind spot" for definitively diagnosing a SAH via LP.

- LP is most sensitive at 12 hours after symptom onset. That's *definitely* not how long we wait to perform our LPs.

Just my $0.02.

Do you have links to this?

I have read it somewhere but ACEP recs are still if CT/LP neg, discharge home.

 

[docB]

I'm intrigued by the idea of using CTA in lieu of LP. I would probably only use this modality if the patient refused the LP but I might put it out as a more viable option. So the question is what do you do with a patient after a negative CTA? Can they go home or do you obs them?

 

[lawj]

What about negative CT within 6 hours of symptom onset? If negative, no need for LP.

Perry JJ et al. Sensitivity of computed tomography performed within six hours of onset of headache for diagnosis of subarachnoid haemorrhage: Prospective cohort study. BMJ 2011 Jul 18; 343:4277.

 

[WilcoWorld]

Perry JJ et al. Sensitivity of computed tomography performed within six hours of onset of headache for diagnosis of subarachnoid haemorrhage: Prospective cohort study. BMJ 2011 Jul 18; 343:4277.

Exactly - if you believe the above article (and I do), then you should recognize that adding a CTA to a negative CT performed within 6 hours of symptom onset does next-to-nada to lower your post-test probability, as it was already very nearly zero.

 

[dotcb]

It's an interesting study, but 1 study prolly shouldn't change practice. I've used it to have discussions with patients about risk and probability. I did do the LP. Story was too good.

We spend a lot of time thinking about and fruitlessly pursuing this diagnosis because it is a rare, can't miss emergency.

 

[WilcoWorld]

I'm not saying that we should skip the LP on the basis of this study alone. I'm saying that this study casts doubt on the value of adding another, more invasive radiographic study to the work up of SAH when you've already got a negative, high quality CT within 6 hours of onset.

 

[FDNewbie]

Similar to how "trouble breathing" means albuterol and lasix? I mean, let's cover everything with blanket terms.

I wasn't advocating for it; merely stating what is.

The elaboration is that catching the bleed early doesn't change outcomes. Treatment of the HTN does. How many head bleeds spend all night in the ICU and get operated on the next day (or never)? And you should be treating the HTN (and pain) before the scan to begin with.

I still think I'd have a hard time justifying even to myself why it took 2-3 hours to see the patient and get a CT scan, only to discover there's a ICH on the CT.

Also, our population is FULL of elderly on coumadin, plavix, pradaxa, and other wonderful drugs. So bleeds are often significant. And maybe I'm still young and overzealous, but I still think every *second* counts, especially when you're at a facility that doesn't have a neurosurgical residcency/fellowship, so you're automatically adding another 30+ min - 1 hr from the time of diagnosis until someone qualified to intervene can actually be in-house and available to intervene.

Are the vast majority of these spontaneous (or HTN-related) bleeds massive and life-threatening? Probably not. But as an EM doc, I like to assume each one IS, *until* proven otherwise.

Oh and just for my education, could you kindly provide a reference for article re: early detection of ICH doesn't change outcome?

People sue for lots of reasons. Rushing people to the scanner (and possibly bumping others out of line) isn't going to stop that

I think it IS good CYA medicine. What can be perceived as an unnecessary delay in diagnosis and/or treatment is VERY high risk IMO.

Find me some that doesn't say that 5% of the population doesn't have an identifiable aneurysm (that isn't bleeding), and that we don't cause harm to people by putting them in a "CT every 6 months for life" group, then I'll buy it.

I think there's a difference in understanding the purpose of the CTA. It isn't (only) to diagnose an aneurysm; the article was (also) specifically looking at "the sensitivity of CT for diagnosing acute SAH" in it of itself. Ie, if I give a contrast load, and there's a break in a vessel somewhere, can I catch it on CT, AS GOOD AS catching it on LP. Remove the entire premise of an aneurysm at all.

CTA does help show where the bleed is, after you've identified a bleed. That's something the neurosurgeon cares about. Not something that matters to us. However, I'm more likely to let concede with a patient about not getting an LP if they agree to a CTA. Similarly, there are a lot of docs who want to do CTAs so they don't have to be doing LPs.

Again, I think we're looking at it differently. There are studies trying to prove that a CTA is JUST AS GOOD at *identifying* the bleed. It's sufficiently sensitive enough to pick up a bleed that a non-con CT would miss. Ie, it's equivalent to an LP. Does that make sense?

I've never gotten an LP in someone less than 2 hours after symptom onset. Hell, they're usually still in the lobby at that point. I would bet more of them fall in the after 12 hour window than under 2 hours.

I guess our experiences are different.

There's also no data for not doing an LP in those patients. Sure, there is a "consensus" that isn't driven by science, but no hard facts. Of course, knowing how reticent everyone is for doing anything with a patient on Plavix, I'm sure they probably didn't even know she was on it (or she said something inane like "something for hypertension")

Which again, is why I'm not saying "everyone change their practice." I'm merely saying that it looks like the CTA sensitivity is approaching - if not surpassing - LP sensitivity, and it may very well become an alternative if not the standard of care if/when available. But again, we'll see what the final verdict is as more data emerges.

 

[FDNewbie]

Do you have links to this?

I have read it somewhere but ACEP recs are still if CT/LP neg, discharge home.

Unfortunately it's not specifically referenced in the emedicine article, but here it is nonetheless: http://emedicine.medscape.com/article/1164341-workup#aw2aab6b5b3. 2nd paragraph.

And yes, I think emedicine is pretty much spot-on in the vast majority of articles/reference material it lists, when it comes to EM diagnosis, workup, and management.

 

[Aesculapius]

There seems to be a lot of sturm and drang about this. Can't you just do the CT, explain to them that there is still a nonzero risk (using whatever number you deem fit from the literature) of a SAH and that you advise a confirmatory LP, and if they refuse, just document that? Or do you still retain liability? It doesn't seem worth it to miss the diagnosis/get a bad outcome/get sued by cutting corners, even if it isn't really clear at this point if you are always actually cutting a corner (if you do CT within 6 hours).

 

[Rendar5]

I'm merely saying that it looks like the CTA sensitivity is approaching - if not surpassing - LP sensitivity, and it may very well become an alternative if not the standard of care if/when available. But again, we'll see what the final verdict is as more data emerges..

How can you surpass a gold standard test? unless there's a gold standard different than the LP?

My honest question is what do you do when you see a 5mm aneurysm. Do you then LP it to see if it's bleeding? do you just admit it? do you refer to Neurosurgery. What do you do if you find the 5mm aneurysm and the LP is negative? Where do you f/u?

 

[FDNewbie]

How can you surpass a gold standard test? unless there's a gold standard different than the LP?

Just like we bypass all sorts of gold standard tests.

What's the gold standard test for diagnosing a PE? It's pulmonary angiography. When was the last time you ordered one? Wait...scratch that...*heard of one* being done? We've now concluded that CTA is just as good for *clinically significant* PEs (not these small subsegmental ones that aren't clinically significant).

I can list a whole HOST of diagnoses where the gold standard is ALWAYS bypassed 99.99% of the time in every ED in the country, because the gold standard is very cumbersome, invasive, and time-consuming. But most importantly, because another NON-gold standard test's sensitivity has approached the acceptable sensitivity level and false negative rate of that gold standard.

My honest question is what do you do when you see a 5mm aneurysm. Do you then LP it to see if it's bleeding? do you just admit it? do you refer to Neurosurgery. What do you do if you find the 5mm aneurysm and the LP is negative? Where do you f/u?

If I have a negative non-con head CT, and for whatever reason I've decided to follow with a CTA, which is also negative, and the patient clinically is asymptomatic/without deficits/neurologic findings, I'm calling the Neurosurgeon specifically for them to tell me to send them home with 1-2 wk outpt F/U, unless the Neurosurgeon wants to admit them. Why?

1) Because I'm pretty confident a CTA would pick up a bleed if it was there. So negative CTA is REALLY negative in my book

2) Because I practice CYA medicine (unfortunately), and want to document the time of the convo & the name of the neurosurgeon. I'm not making an independent decision on something like that. When in doubt, load the boat. I've consulted the specialist, and am following his recs.

3) Because an incidental finding of an aneurysm ISN'T my concern. I didn't do the CTA to find an aneurysm. I did the CTA to CONFIRM that 5% chance there's a bleed and the NON-con CT missed it. So I'm ONLY looking for bleeding. Anything else I find is an incidentaloma, which I'd dispo as appropriate (in this case with a call to Neurosurg, to look over the CT and provide their recs, which I'd document in the chart and carry out).

Now I'm not saying this is the right or only way to do things. I'm merely answering your question. So I'd love to hear what you guys think of that approach, the possible pitfalls, etc.

 

[neusu]

1) Because I'm pretty confident a CTA would pick up a bleed if it was there. So negative CTA is REALLY negative in my book

2) Because I practice CYA medicine (unfortunately), and want to document the time of the convo & the name of the neurosurgeon. I'm not making an independent decision on something like that. When in doubt, load the boat. I've consulted the specialist, and am following his recs.

3) Because an incidental finding of an aneurysm ISN'T my concern. I didn't do the CTA to find an aneurysm. I did the CTA to CONFIRM that 5% chance there's a bleed and the NON-con CT missed it. So I'm ONLY looking for bleeding. Anything else I find is an incidentaloma, which I'd dispo as appropriate (in this case with a call to Neurosurg, to look over the CT and provide their recs, which I'd document in the chart and carry out).

1) In aneurysmal sah, the aneurysm does not continue to bleed. CTA would merely show presence or absence of an aneurysm. We still admit LP positive CTA negative patients for an angiogram and keep them for a week to repeat the angiogram if it's truly a good story. If the aneurysm were still bleeding (an active cerebral arterial bleed), the patient would decompensate and die.

2) What? Why? For acute headache?

3) Again, CTA is not to show bleeding, it's to show presence of an aneurysm. An incidental aneurysm in the setting of an acute headache may be at risk of rupture (thrombosed, suddenly expanded etc.), but the test for bleeding is an LP. If the patient refuses you're not selling it hard enough or the story isn't good. On a similar note, if the patient refuses an LP to diagnose a disease that will inevitably need surgery or interventional surgery, do they really want treatment anyway?

 

[Rendar5]

1) If the patient refuses you're not selling it hard enough or the story isn't good. On a similar note, if the patient refuses an LP to diagnose a disease that will inevitably need surgery or interventional surgery, do they really want treatment anyway?

There are plenty of people who'll refuse the prospect of a needle in the back, no matter how hard you push. They'll say they want treatment in retrospect though

 

[FDNewbie]

1) In aneurysmal sah, the aneurysm does not continue to bleed. CTA would merely show presence or absence of an aneurysm. We still admit LP positive CTA negative patients for an angiogram and keep them for a week to repeat the angiogram if it's truly a good story. If the aneurysm were still bleeding (an active cerebral arterial bleed), the patient would decompensate and die.

Yes, and I've had continued arterial bleeding, which resulted in continued decompensation. In some cases, they were saved, and in some cases, they herniated before anyone was available in-house to intervene. And in some cases they were already so forgone it was a lost cause. Where you are and how early you are in detecting that is one of the points they're arguing a CTA is sensitive for (detecting a bleed). That's not my assumption; that's what the article is stating, as a distinct and discrete ADDITIONAL point to it's ability to detect aneurysms. Two distinct entities.

2) What? Why? For acute headache?

Yes. you just answered your question in your Point #3 below. It's an aneurysm. It's high risk (how high isn't my concern; it's high risk period). So if there's no bleed, but there's an aneurysm, part of the BEST CYA things you can do is establish follow-up yourself. That shows you've gone above & beyond the call of duty for your patient.

3) Again, CTA is not to show bleeding, it's to show presence of an aneurysm. An incidental aneurysm in the setting of an acute headache may be at risk of rupture (thrombosed, suddenly expanded etc.), but the test for bleeding is an LP. If the patient refuses you're not selling it hard enough or the story isn't good.

On a similar note, if the patient refuses an LP to diagnose a disease that will inevitably need surgery or interventional surgery, do they really want treatment anyway?

Yep. There are plenty of geniuses who deny themselves essential life-saving therapy. Doesn't mean they don't need it though.

 

[deleted109597]

You're still the only one here endorsing CTA to identify bleeding. You're actually the only one I've ever heard of saying it. CTA is to identify the aneurysm, not the bleeding.

 

[FDNewbie]

You're still the only one here endorsing CTA to identify bleeding. You're actually the only one I've ever heard of saying it. CTA is to identify the aneurysm, not the bleeding.

I'm not endorsing anything. I'm saying there's early data to suggest that this may be proven in near future to be as good as CT + LP. And if I'm the only person saying it, that's sad, cuz I must be the only person reading the literature.

1) Computed Tomographic Angiography for the Evaluation of Aneurysmal Subarachnoid Hemorrhage. Acad Emerg Med. 2006;13:486-492. (Prospective; 116 patients). http://www.medschool.lsuhsc.edu/emergency_medicine/docs/CTA.pdf

Conclusions: In this pilot study, CTA was found to be useful in the detection of cerebral aneurysms and may be useful in the diagnosis of aneurysmal SAH. A larger multicenter study would be useful to confirm these results.

2) Computed tomographic angiography as the primary diagnostic study in spontaneous subarachnoid hemorrhage. (I don't think the title gets more obvious than that homie). J Neuroimaging. 2007 Oct;17(4):295-9. http://www.ncbi.nlm.nih.gov/pubmed/17894616

CONCLUSION: CTA can provide prompt and accurate diagnostic and anatomic information in the setting of SAH with an excellent detection rate in acute ruptured aneurysms. These findings suggest an increased role for CTA in the evaluation of cerebral aneurysms.

3) Can computed tomography angiography of the brain replace lumbar puncture in the evaluation of acute-onset headache after a negative noncontrast cranial computed tomography scan? (title is self-explanatory). Acad Emerg Med. 2010 Apr;17(4):444-51. http://www.ncbi.nlm.nih.gov/pubmed/20370785

CONCLUSIONS: CT followed by CTA can exclude SAH with a greater than 99% posttest probability. In ED patients complaining of acute-onset headache without significant SAH risk factors, CT/CTA may offer a less invasive and more specific diagnostic paradigm. If one chooses to offer LP after CT/CTA, informed consent for LP should put the pretest risk of a missed aneurysmal SAH at less than 1%.

And from EBMedicine.net ( http://www.ebmedicine.net/topics.php?paction=showTopic&topic_id=193):

"As of 2009, although the results are promising for CTA in the diagnosis of ruptured aneurysm, the current diagnostic strategy for SAH cannot be altered at this time without larger prospective trials examining test characteristics, risks, benefits, and the cost-effectiveness of this strategy."

And that's ALL I'm trying to say.

 

[neusu]

I'm not endorsing anything. I'm saying there's early data to suggest that this may be proven in near future to be as good as CT + LP. And if I'm the only person saying it, that's sad, cuz I must be the only person reading the literature.

And that's ALL I'm trying to say.

I read the aforementioned studies, none of which indicate they are advocating use of CTA to diagnose subarachnoid hemorrhage. You seem to be misunderstanding the premise and shifting the argument. The studies support using CTA as a modality to identify an aneurysm and each one indicates that the aneurysm may not have bleed. Mere presence of an aneurysm in the setting of a headache, especially the most common which are also at the lowest risk of rupture (anterior circulation and < 7mm: 0% 5 year risk), does not warrant any intervention. You are shifting the question from "is there blood in the head?" to "is there an aneurysm?" which is a dangerous paradigm shift to make as the risk of re-rupture (60% fatal) in the first 2 weeks is 20%, especially when CTA is weak at detecting aneurysms < 3mm.

 

[Rendar5]

I'm actually familiar with a couple of those studies. Unfortunately don't have time to fully review the latter 2, but I can say that study 2 doesn't even answer the question proposed here, study 3 is mathematical masturbation, not an actual study, and study 1 kind of shows the unintended consequences of CTA.

In the first one, CTA over-identified aneurysms. in the case that CT was negative, the LP identified two bleeds, CTA caught those 2, plus 3 more aneurysms in non-SAH patients. If you avoided the LP and just used CTA, you would've been left in the dark with a 40% chance that it as a SAH and a 60% chance that it was not. Thus, LP was needed regardless of CTA results.

In addition those 3 non-SAH cases all underwent traditional angiography, a procedure that is probably riskier than an LP. 1 of those cases ended up being a false-positive CTA (thus, unnecessary risk and lots of added expense) the other 2 cases were given brain surgery for what may have been incidentilomas of a sort. These neurosurgeries would certainly not have been done if the ED did not perform a CTA.

As for the second study, despite your smart-aleck claim that the title said it all, CTA was not even tested as a diagnostic tool. It was a study to evaluate it's use as a presurgical evaluation.

The third study (if it's the one I recall, not sure it is) while some people swear by it, it really doesn't make much sense to me. It's a mathematical model that somehow equates catching an aneurysm (CTA) to catching a bleed, and then adding that into those bleeds caught by CT. Unfortunately, it provides no real-world data and may not even be logically sound, as I explained above. I'd have to go into the study though to see if it makes the leap of faith logic that I think it does.

I'm not endorsing anything. I'm saying there's early data to suggest that this may be proven in near future to be as good as CT + LP. And if I'm the only person saying it, that's sad, cuz I must be the only person reading the literature.

1) Computed Tomographic Angiography for the Evaluation of Aneurysmal Subarachnoid Hemorrhage. Acad Emerg Med. 2006;13:486-492. (Prospective; 116 patients). http://www.medschool.lsuhsc.edu/emergency_medicine/docs/CTA.pdf

Conclusions: In this pilot study, CTA was found to be useful in the detection of cerebral aneurysms and may be useful in the diagnosis of aneurysmal SAH. A larger multicenter study would be useful to confirm these results.

2) Computed tomographic angiography as the primary diagnostic study in spontaneous subarachnoid hemorrhage. (I don't think the title gets more obvious than that homie). J Neuroimaging. 2007 Oct;17(4):295-9. http://www.ncbi.nlm.nih.gov/pubmed/17894616

CONCLUSION: CTA can provide prompt and accurate diagnostic and anatomic information in the setting of SAH with an excellent detection rate in acute ruptured aneurysms. These findings suggest an increased role for CTA in the evaluation of cerebral aneurysms.

3) Can computed tomography angiography of the brain replace lumbar puncture in the evaluation of acute-onset headache after a negative noncontrast cranial computed tomography scan? (title is self-explanatory). Acad Emerg Med. 2010 Apr;17(4):444-51. http://www.ncbi.nlm.nih.gov/pubmed/20370785

CONCLUSIONS: CT followed by CTA can exclude SAH with a greater than 99% posttest probability. In ED patients complaining of acute-onset headache without significant SAH risk factors, CT/CTA may offer a less invasive and more specific diagnostic paradigm. If one chooses to offer LP after CT/CTA, informed consent for LP should put the pretest risk of a missed aneurysmal SAH at less than 1%.

And from EBMedicine.net ( http://www.ebmedicine.net/topics.php?paction=showTopic&topic_id=193):

"As of 2009, although the results are promising for CTA in the diagnosis of ruptured aneurysm, the current diagnostic strategy for SAH cannot be altered at this time without larger prospective trials examining test characteristics, risks, benefits, and the cost-effectiveness of this strategy."

And that's ALL I'm trying to say.

The 3rd study is simply a mathematical model. It's not an actual study.

 

[iridesingltrack]

As an ED physician I understand the desire to wish/want/hope for CTA to be as good as LP but...

I am also a neurocritical care fellow and in that regard I have seen several CTA negative, angio positive aneurysmal hemorrhages. Yes, there are some perimesencephalic hemorrhages but if the story is concerning for SAH, and the CT is negative then the patient probably should get an LP...for now anyway.

As to the Perry article...Personally, I like it; It hasn't changed my practice per se, but it has modified it, which, I guess ultimately is a change.

iride

 

[dotcb]

This debate seems analogous to the arguments of cardiac CT for ACS. They say the negative predictive value and follow-up outcomes are good. The problem is that while cardiac CT is good at diagnosing CAD, it does not tell you whether there is an unstable plaque that is going to rupture. And when we can already rule out STEMI with EKG, and NSTEMI with troponins, does this add a great deal?

And while CTA brain is good for aneurysm determination, it doesn't answer whether there is ICH, which is what LP does.

 

[deleted109597]

I'm not endorsing anything. I'm saying there's early data to suggest that this may be proven in near future to be as good as CT + LP. And if I'm the only person saying it, that's sad, cuz I must be the only person reading the literature.

Multiple useless studies

And that's ALL I'm trying to say.

And all we are trying to say is that you're wrong, and being obtuse about it. Sure, one day we might be able to. However, the radiation and contrast risks, coupled with the multiple misses, makes it a bad idea to do it now. We are trying to educate lots of people here, you included. And I'm telling you right now that if you CTA and miss a bleed, you're going to lose a lawsuit as it isn't the standard of care.
Pulmonary angiograms are no longer the standard of care, no matter what your old Harrison's tells you. The brain vasculature is much smaller and thus you would need much thinner slices to get the same level of detail.
And finally, while you may have some anecdotal "patients who were saved" by CTA, whatever that means, it still isn't as good as an LP. All people are trying to do is get out of the LP, and I find that bad medicine.
(And yes, for the record, I have in the past advocated CT+CTA as better when you let someone leave AMA after refusing LP than simple CT)

 

[FDNewbie]

And all we are trying to say is that you're wrong, and being obtuse about it. Sure, one day we might be able to. However, the radiation and contrast risks, coupled with the multiple misses, makes it a bad idea to do it now. We are trying to educate lots of people here, you included. And I'm telling you right now that if you CTA and miss a bleed, you're going to lose a lawsuit as it isn't the standard of care.

Ninja, did I say anywhere that I think it IS the standard of care? I simply said I think it MAY become it, or be an equivalent alternative to LP *in the near future.* So we don't disagree at ALL on this. If I misrepresented my point, my bad. But I thought it was pretty clear.

Re: the radiation risks, I'm not looking to start another debate, but for every radiation risk article, there's another that pops up saying there's no real increased CA risk with repeated radiation. Is that cuz the radiologists and imaging lobby is very strong & well funded? Or is it just a matter of time when all the ppl we've CTed their abdomens 17 times a month for their "abdominal migraines" grow old enough and we see the long-term effects of making them glow in our EDs... I'm curious to see.

And finally, while you may have some anecdotal "patients who were saved" by CTA, whatever that means, it still isn't as good as an LP.

My point in that respect was in getting the imaging PROMPTLY. Had nothing to do with the CTA. I was stating I actually like getting the head CT immediately upon arrival based on a good story, cuz then I'm on the phone with Neurosurgery before the patient has even come back FROM the CT scanner. With significant bleeds, I think every second counts.

You, however, had mentioned that there's no benefit to diagnosing ICH in under 3 hours (or something to that effect). I was asking you for your reference, because bad bleeds can herniate WELL prior to 3 hours. So catching them as early as possible seems like the logical thing to do.

All people are trying to do is get out of the LP, and I find that bad medicine.

That's definitely not my goal. I'm a procedure freak, and don't even try to hide it. But I'm interested to see if CTA will prove to be equivalent to LP in the near future. I think it would be very beneficial, esp. when not all EM physicians are comfy doing invasive procedures, or good at it at that.

(And yes, for the record, I have in the past advocated CT+CTA as better when you let someone leave AMA after refusing LP than simple CT)

I'll def. agree to that. You've done the best you can to evaluate the patient, within the constraints the patient has placed on you.

 

[neusu]

You, however, had mentioned that there's no benefit to diagnosing ICH in under 3 hours (or something to that effect). I was asking you for your reference, because bad bleeds can herniate WELL prior to 3 hours. So catching them as early as possible seems like the logical thing to do.

SAH and ICH are two entirely different entities. In SAH, the bleed is arterial, transient and the diagnostics are to correlate the presenting symptoms of HA, hemiparesis, stupor, and/or coma with a pathology. Noncontrast CT, MRI, and LP are the diagnostic procedures for SAH. If a patient is even a candidate for an LP there is realistically no way an expanding hematoma with impending herniation is possible. To humor your point, true, an aneurysm can re-rupture. A CTA in this scenario would neither predict for, nor catch the event in any meaningful way greater than a repeat noncontrast CT brain. An ICH, on the otherhand, is diagnosed with noncontrast head CT and is prone to expansion. A repeat noncontrast CT brain at an interval to assess expansion is typically conducted, and if there is a focal neurologic exam change in the interim the repeat CT is expedited. A patient herniating in under 3 hours undoubtedly would have progressive exam changes. FWIW, other active arterial bleeds such as epidural hematoma (EDH) are constrained by fixed structures such as the cranial vault and the dura and do not expand at the same rate as an unsecured active aneurysm bleed. An aneurysm bleeds in to the subarachnoid space, stops bleeding, and is at high risk of rebleeding. If the initial bleed doesn't stop there is a large amount of SAH on the initial CT. If the patient rebleeds, there is a sudden deterioration and more than likely herniation is not the issue but instead hydrocephalus from obstructive intraventricular hemorrhage.

 

[FDNewbie]

SAH and ICH are two entirely different entities. In SAH, the bleed is arterial, transient and the diagnostics are to correlate the presenting symptoms of HA, hemiparesis, stupor, and/or coma with a pathology. Noncontrast CT, MRI, and LP are the diagnostic procedures for SAH. If a patient is even a candidate for an LP there is realistically no way an expanding hematoma with impending herniation is possible. To humor your point, true, an aneurysm can re-rupture. A CTA in this scenario would neither predict for, nor catch the event in any meaningful way greater than a repeat noncontrast CT brain. An ICH, on the otherhand, is diagnosed with noncontrast head CT and is prone to expansion. A repeat noncontrast CT brain at an interval to assess expansion is typically conducted, and if there is a focal neurologic exam change in the interim the repeat CT is expedited. A patient herniating in under 3 hours undoubtedly would have progressive exam changes. FWIW, other active arterial bleeds such as epidural hematoma (EDH) are constrained by fixed structures such as the cranial vault and the dura and do not expand at the same rate as an unsecured active aneurysm bleed. An aneurysm bleeds in to the subarachnoid space, stops bleeding, and is at high risk of rebleeding. If the initial bleed doesn't stop there is a large amount of SAH on the initial CT. If the patient rebleeds, there is a sudden deterioration and more than likely herniation is not the issue but instead hydrocephalus from obstructive intraventricular hemorrhage.

Actually, that's not correct. ICH is a blanket term that refers generally to pretty much ANY bleeding from the cranium down. SAH is a TYPE of ICH (extra-axial, yes, but we're not gonna get into that).

Specific to SAH, I've had severe SAHs in old patients that resulted in herniation after several hours (2-3). My point was in reference to Ninja stating there's no difference in whether you diagnose it in the first 3 hours or so. I'm having trouble understanding how that's the case, when even the initial bleed can be life-threatening.

This has NOTHING to do with LPs. I had mentioned our EMS protocol was to proceed directly to the CT scanner. He said that's essentially not the way to do it; an H&P first is the way to go = delay EASILY 30 min - 1 hr if not more before you even put in the head CT order (if not before because of clinical deterioration in the meantime).

So again, my point was simple. Delayed imaging in head bleeds is bad. And I'd have a VERY hard time understanding how there's no change in morbidity or mortality whether you diagnose it within 1 or 3 hours (as he had stated studies have shown), when I've seen patients DIE in that time period.

So again, looking for a reference on that. So I can learn too.

 

[neusu]

Actually, that's not correct. ICH is a blanket term that refers generally to pretty much ANY bleeding from the cranium down. SAH is a TYPE of ICH (extra-axial, yes, but we're not gonna get into that).

Actually, that's not correct. ICH (intracerebral hemorrhage) is a blanket term for any intraaxial bleeding, or within the parenchyma of the brain (intraparenchymal, intraventricular, etc.). Extraxial hemorrhages include subarachnoid, subdural, epidural and so forth. You, perhaps, are confusing the "C" in ICH (meaning cerebral) for cranial.

Specific to SAH, I've had severe SAHs in old patients that resulted in herniation after several hours (2-3). My point was in reference to Ninja stating there's no difference in whether you diagnose it in the first 3 hours or so. I'm having trouble understanding how that's the case, when even the initial bleed can be life-threatening.

Again, I'm not sure if you are confused, but from the sounds of your description the "old patients" likely had a traumatic SAH which has nothing to do with the discussion of using CTA for management of an aneurysmal SAH.

This has NOTHING to do with LPs. I had mentioned our EMS protocol was to proceed directly to the CT scanner. He said that's essentially not the way to do it; an H&P first is the way to go = delay EASILY 30 min - 1 hr if not more before you even put in the head CT order (if not before because of clinical deterioration in the meantime).

So again, my point was simple. Delayed imaging in head bleeds is bad. And I'd have a VERY hard time understanding how there's no change in morbidity or mortality whether you diagnose it within 1 or 3 hours (as he had stated studies have shown), when I've seen patients DIE in that time period.

I won't fight McNinjas fight regarding the timing of imaging, but his post seems to be when he mentioned "hemorrhagic CVA," he is discussing hemorrhagic stroke, or ICH, and not aneurysm/SAH. Regardless of the pathology, a detailed history is pertinent. If it's a SAH we park them in the neuroICU until the OR or biplane suite is free and if it's a CVA we park them in the neuroICU and do serial neuro exams until a deterioration or an interval repeat CT is obtained.

As for your "old patient" who herniates in the ED (be it hemorrhagic stroke or traumatic SAH), undoubtedly an associated coagulopathy must be reversed which a) takes time and b) needs to be identified either by history or laboratory studies. Even then, the etiology plays a big role on the operative discussion i.e. brainstem, cerebellar, amyloid, thalamic bleeds tend to lead to goals of hospitalization discussion rather than a book an OR STAT.

 

[FDNewbie]

Actually, that's not correct. ICH (intracerebral hemorrhage) is a blanket term for any intraaxial bleeding, or within the parenchyma of the brain (intraparenchymal, intraventricular, etc.). Extraxial hemorrhages include subarachnoid, subdural, epidural and so forth. You, perhaps, are confusing the "C" in ICH (meaning cerebral) for cranial.

Yea I was using it for intracranial. Is that not the correct/proper abbreviation? I've seen both (cerebral and cranial) used.

Again, I'm not sure if you are confused, but from the sounds of your description the "old patients" likely had a traumatic SAH which has nothing to do with the discussion of using CTA for management of an aneurysmal SAH.

I agree that it has nothing to do with it. And etiology is hard to discern, but I can distinctly recall a few cases where we didn't have a history of a fall; they were sitting in a chair/desk etc, and someone noted a sig. metal status change in them. Brought them in, BAD SAH.

I won't fight McNinjas fight regarding the timing of imaging, but his post seems to be when he mentioned "hemorrhagic CVA," he is discussing hemorrhagic stroke, or ICH, and not aneurysm/SAH. Regardless of the pathology, a detailed history is pertinent. If it's a SAH we park them in the neuroICU until the OR or biplane suite is free and if it's a CVA we park them in the neuroICU and do serial neuro exams until a deterioration or an interval repeat CT is obtained.

Yea I figured as much. And that's what my last few posts were in reference to, and how - IMO - regardless of the etiology, it seems that timing is everything.

As for your "old patient" who herniates in the ED (be it hemorrhagic stroke or traumatic SAH), undoubtedly an associated coagulopathy must be reversed which a) takes time and b) needs to be identified either by history or laboratory studies. Even then, the etiology plays a big role on the operative discussion i.e. brainstem, cerebellar, amyloid, thalamic bleeds tend to lead to goals of hospitalization discussion rather than a book an OR STAT.

Agreed.

 

[deleted109597]

Yea I figured as much. And that's what my last few posts were in reference to, and how - IMO - regardless of the etiology, it seems that timing is everything.

And yet, your opinion is still wrong.
Sorry.

 

[ptarmigan]

FD, I think you should strongly reconsider using CTA to rule out spontaneous SAH. I have never heard of anyone doing that. Ask your colleagues if they are doing that. It is certainly not the standard of care or even an idea I have ever heard suggested by anyone. This could get you into big trouble because in the courtroom you will be judged by the standard of care. You are better off with a non con CT head and a good MDM note about why your suspicion is low.

If you are looking to rule out spontaneous SAH, you need to do an LP and that is non-negotiable. If you are looking for an aneurysm, then fine do a CTA. A patient without an aneurysm is extremely unlikely to have a spontaneous SAH, and that is where it could help you. This algorithm for working up headaches is also not the standard of care, but at least it actually makes some sense.

Do not call a neurosurgeon to dispo a headache. Call a neurosurgeon when you need to go to the OR. Documenting in the chart that the patient is going to follow up with a neurosurgeon in 3 days will not help you in the slightest if the patient drops dead of spontaneous SAH tomorrow and you did not do an LP. Remember that in acute headache with GCS 15, the LP is to catch the sentinel bleed, not the bleed that is going to kill them. And your neurosurgeons will hate you. They like to operate. They do not like talking to anyone over the phone and they do not like seeing patients in clinic. In general, they do not like interacting with people that have a GCS greater than 8.

 

[southerndoc]

This could get you into big trouble because in the courtroom you will be judged by the standard of care.

In some states (such as mine), an emergency provider must be grossly negligent and not just fail to meet the standard of care before a patient can get a lawsuit certified to go forward.

Do not call a neurosurgeon to dispo a headache. Call a neurosurgeon when you need to go to the OR. Documenting in the chart that the patient is going to follow up with a neurosurgeon in 3 days will not help you in the slightest if the patient drops dead of spontaneous SAH tomorrow and you did not do an LP. Remember that in acute headache with GCS 15, the LP is to catch the sentinel bleed, not the bleed that is going to kill them. And your neurosurgeons will hate you. They like to operate. They do not like talking to anyone over the phone and they do not like seeing patients in clinic. In general, they do not like interacting with people that have a GCS greater than 8.

I'm glad I don't work at a place where our neurosurgeons are bothered about consults. I'm on a first name basis with all of them, and I sometimes call for guidance on things and I've never had one badmouth me for doing so.

 

[ptarmigan]

Wow, that's nice. How is gross negligence different from negligence? Unfortunately in my state the standard is plain old negligence which is just another way of saying breach of the standard of care.