CT Scan before LP

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

EM Guy

Junior Member
10+ Year Member
15+ Year Member
Joined
Jan 28, 2005
Messages
20
Reaction score
0
Most of what I've been able to find on this issue seems to show that you DON'T necessarily need to scan every head prior to LP (although this seems to be the default standard of care at my hospital). In general, it seems like most people follow the guidelines from the NEJM article from the 2005 LLSA list. How are the rest of you practicing?

Members don't see this ad.
 
It depends on what you're looking for.

If you're looking for SAH, then yeah do the CT first because you can find it there, it's less invasive and can be faster.

If you're looking for meningitis, then you can skip the CT if you meet all these criteria (and I'm sure I'm going to miss one because I don't have the article in front of me):
1. Normal neuroloic and mental status exam
2. Less than 65 years old
3. Not immunocompromised
4. Normal fundoscopic exam

It's #4 that gets me everytime. I suck at fundoscopic exams and could never be sure I had seen a normal optic disc even if I had a good look.
 
Members don't see this ad :)
From extensive review of the literature the proverbial fear of herniation is one of those things of mythos. If you actually look at the literature, it is incredibly sketchy.

In general, most of the attendings here (when talking about meningitis, not SAH) if there is nl nuerologic/ms exam will do the LP without head CT.

For SAH, I scan first because if its there, I don't have to LP
 
roja said:
From extensive review of the literature the proverbial fear of herniation is one of those things of mythos. If you actually look at the literature, it is incredibly sketchy.

In general, most of the attendings here (when talking about meningitis, not SAH) if there is nl nuerologic/ms exam will do the LP without head CT.

For SAH, I scan first because if its there, I don't have to LP

Not true, ask any old time neurosurgeon. You just don't see it anymore because the CAT scans are readily available. Medical lit in the 70s and before was always sketchy, in fact it was badly written. Doesn't mean it didn't describe a real phenomenon

I've personally herniated a 21 y/o female with delerium, AMS, fever, a whited out lung and no focal neuro findings. Didn't have a scanner at Baltimore City in '78, would have taken 5 hours to send her to JHH. She had a symmetrical hydrocephalus. The pneumonia was an aspiration secondary to the CNS problem. We got her out of it with an emergency twist drill, but it's not a situation you want to be in.

I've also seen two small children herniate after tap (they had midbrain tumors). One didn't make it.

I think that the incidence of bacterial meningitis is going down, while the incidence of viral CNS infection and brain mass or abscess is the same or higher.

Probably USC diver's criteria are pretty safe, but why not get the study? Even if there is no risk of herniation, it often gives you a lot of information that you wouldn't get with just the tap (like a mass that hasn't raised the pressure much yet and won't affect the CSF findings.)

Here's how I do it:

1. Acute Meningitis Syndrome (fever and AMS). Give them antibiotics, get the CT, get the tap. If they have bacterial meninigitis its very unlikely their CNS results will be equivocal on the basis of some anitibiotics given shortly before.
NOte that this is a small group consisiting of 1/3 of allABM, Cerebral abscess and oSepsis elsewhere

2. Subacute meningitis syndrome (fever and HA, normal neuro) Send them for the CT, do the tap, antibiotics if necessary. Larger group consisitng of 2/3 of all ABM, most viral infections and simple non-cns infections.

3. Encephilitis syndrome (AMS, focal findings, +/- fever). CT or MRI, tap if safe and indicated. Consists of most encepahlitis, brain abscesses, SAH and many other CNS conditions

The CAT scanner is your friend.
 
In most Ontario hospitals, CT scanning of the head is done before a suspected meningitis LP with any of the following: (taken from a neurology handbook):

CT scanning of the head before LP in suspected meningitis:

- The most worrisome contraindication to lumbar puncture is the suspicion of increased ICP due to a cerebral mass lesion. Performing a lumbar puncture in these patients may lead to either trans-tentorial or uncal herniation and acute neurological deterioration.

- The absence of all these features makes a significant lesion precluding LP very unlikely:

1) Age > 60 yrs

2) Immuno-compromised state

3) History of CNS disease (eg. grand mal seizures, brain tumour, hydrocephalus, multiple sclerosis)

4) Seizure within one week of presentation

5) Abnormal LOC

6) Unable to answer two questions correctly or follow two commands

7) Abnormal neurological examination (visual field defect, facial palsy, pronator drift, aphasia)

eg. An LP may safely be performed without first doing a CT head in a young previously healthy patient with no history of seizures, a normal level of consciousness and a normal neurological exam.
 
There is an ACEP Clinical Policy on whether or not you need a CT before tapping.

Basically, if they have:
signs of incr ICP (papilledema, which we are all great at looking for)
AMS
focal neuro deficit
or absent venous pulsations on fundo exam (cant tell you the last time I noticed that)
then CT scan before is indicated

Otherwise, tap away, my friends. Or so says ACEP

Q
 
How sensitive is a CT scan for increased ICP? I thought the literature states that the pressure can be elevated and have a completely normal CT scan, is it very valuable then? Anyone know the sensitivity?
 
Unless you plan on dilating the eyes every time, looking for papiledema is really insensitive. During my optho elective I asked over a dozen opthalmology attendings if they would ever be comfortable commenting on papiledema without first dilating the eyes. They all uniformly said absolutely not. These are the guys who make a living out of looking at eyes.
Additionally, by those criteria, you would essentially HAVE TO see venous pulasations to be able to skip over scanning. As you know, almost half of the population do NOT have visible venous pulsations (and this is with dilated exams); and of course absence of venous pulsations is not specific for pathology. Seems like just the requirements for the eye portion of the criteria could be rather time-intensive and not necessarily reduce the overall number of scans that you need to perform.
 
EM Guy said:
Unless you plan on dilating the eyes every time, looking for papiledema is really insensitive. During my optho elective I asked over a dozen opthalmology attendings if they would ever be comfortable commenting on papiledema without first dilating the eyes. They all uniformly said absolutely not. These are the guys who make a living out of looking at eyes.
Additionally, by those criteria, you would essentially HAVE TO see venous pulasations to be able to skip over scanning. As you know, almost half of the population do NOT have visible venous pulsations (and this is with dilated exams); and of course absence of venous pulsations is not specific for pathology. Seems like just the requirements for the eye portion of the criteria could be rather time-intensive and not necessarily reduce the overall number of scans that you need to perform.

Plum and Posner in the classic book "diagnosis of stupor and coma" gave their series of patients referred to them for chronic increased ICP. I believe that about 1/2 had signs of papilledema.

Last year, I think somebody suggested you could use tonometry to diagnose ICP in pseudtumor.

In any case, the eye signs are no guarantee of no ICP elevation, particularly acutely.
 
iatrosB said:
How sensitive is a CT scan for increased ICP? I thought the literature states that the pressure can be elevated and have a completely normal CT scan, is it very valuable then? Anyone know the sensitivity?

The issue we're talking about here is presence of a asymetric mass lesion or symmetrical noncommunicating hydrocephalus that can be seen on CT. With a normal CT, the ICP can be elevated (it usually is in meningitis) but herniation is unlikely.
 
I hear different people referring to Plum and Posner's book all the time. Is this still worth reading as it was last updated in 1982? I just hate wading through through tons of material and discarding that which is outdated trying to mine for pearls.
 
BKN said:
The issue we're talking about here is presence of a asymetric mass lesion or symmetrical noncommunicating hydrocephalus that can be seen on CT. With a normal CT, the ICP can be elevated (it usually is in meningitis) but herniation is unlikely.

Ok, I see. So you usually aren't worried about herniation with meningitis? If the CT is normal, there is a low probability of herniation...am I understanding that correctly?
 
EM Guy said:
I hear different people referring to Plum and Posner's book all the time. Is this still worth reading as it was last updated in 1982? I just hate wading through through tons of material and discarding that which is outdated trying to mine for pearls.
Several months ago, BKN referred it to me about a question I had pertaining to increased intracranial pressure. It does a good job explaining a lot of these concepts, although again as a book from 1982 I don't know if any of the concepts were revised or changed in other literature.
 
iatrosB said:
Ok, I see. So you usually aren't worried about herniation with meningitis? If the CT is normal, there is a low probability of herniation...am I understanding that correctly?

got it.

I hear different people referring to Plum and Posner's book all the time. Is this still worth reading as it was last updated in 1982? I just hate wading through through tons of material and discarding that which is outdated trying to mine for pearls.

If you read it selectively, I think it is. You'll get a general classification of coma into four groups: diffuse or metabolic coma (approx 70%), Supratentorial mass lesions causing herniation (20%), Infratentorial destructive lesions (10%) and psychiatric unresponsiveness (few). You'll learn a 5 point exam that will classify the difference between structural and metabolic coma with >95% accuracy in 2 minutes. You'll learn an initial approach and you'll often be able to pinpoint the lesion before the imaging awing and amzing your friends. And you'll relearn the classic understanding of the comatose state.

Obviously you don't read it to learn imaging. The general therapeutic approach hasn't changed that much. Treatment has changed for specific diseases causing coma, but that's not what the book is about.

I bought the only more recent monograph Young's Coma and impaired conciousness in an effort to update my lecture on this. It added a little, but honestly it was very similar and I thought the classic put it better.
 
Top