1. a 0.5ml CSF sample in a fluoride EDTA tube for glucose and protein estimation (to biochemistry lab)
2. Two 5.0ml CSF samples in a plain sterile universal container for cell count and culture (to the microbiology lab)
3. A final 1.0ml CSF sample in a plain universal container for bilirubin spectrophotometry (to the biochemistry lab or neurochemistry lab - whoever does the CSF spectrophotometry). Remember: this guidance is for suspected subarachnoid haemorrhage only.

Having been telling people for years that the order of tubes does not matter, the new guidance is that the spectrophotometry sample should be the third or fourth one collected, and preferably not the first one. The reason for this is that oxyhaemoglobin can obscure bilirubin on spectrophotometry, and oxyHB is most likely to appear in vitro from the first sample where blood from the site of puncture can be caught up with the CSF. This is a new practice.

The other common scenarios are suspected multiple sclerosis or other inflammatory disease - in this case a 2.0ml CSF sample is needed for detection of oligoclonal bands (immunology lab). Rarely CSF cytology is needed in cases of suspected malignant meningitis. CSF cytology needs to be received by the lab within hours and spun down for cytological examination a 2.0ml to 5.0ml CSF sample would be preferable.

Remember that paired serum is needed for GLUCOSE, PROTEIN, BILIRUBIN, OLIGOCLONAL BANDS.

Here’s the all encompassing list:

1. 0.5ml CSF in fluoride EDTA container
2. 5.0ml CSF in plain container for cell count
3. 5.0ml CSF in plain container for culture
4. 2.0ml CSF in plain container for spectrophotometry
5. 2.0ml CSF in plain container for oligoclonal bands to immunology
6. 2.0ml CSF in plain container for CSF cytology to neuropathology
7. 2.0ml CSF in plain container for viral PCR (polymerase Chain reaction - suspected herpes simplex encephalitis).

Total CSF collected is 18.5ml, which is about 50% of the contents of the lumbar cistern. It will take about 50minutes for 18mls of CSF to be replaced by normal rates of CSF synthesis.

Personally I think that this is probably a bit excessive. The 5ml samples could be reduced to 2mls without seriously compromising analysis. For an emergency LP - typically for acute headache or suspected CNS infection, I would compromise to:

1. 0.5ml CSF in fluoride EDTA container - glucose and protein
2. 2.0ml CSF in plain container for cell count and culture
3. 2.0ml CSF in plain container for spectrophotemetry

PLEASE LABEL this sample specially for spectrophotometry, as it is least likely to be contaminated by traumatic blood.

4. 2.0ml CSF in plain container for viral PCR (polymerase Chain reaction - suspected herpes simplex encephalitis).

CSF cytology could be requested if malignant meningitis is suspected clinically (this is very rare), and oligoclonal bands are rarely needed acutely, but could be useful if confusion cases where CNS inflammation could be a rare cause.

Like a lot of medicine a cookbook list to cover all eventualities is not really possible unless you, your patient and the labs are happy to get 16.5mls or more of CSF each time an LP is performed. My main worry is that an agitated patient or someone in whom needle placement is difficult will not allow enough time to get 18mls of CSF collected. In a scenario where CSF collection is difficult, you may need to make a professional judgement as to which tests would have priority.

The other important issue is to PROTECT YOUR SAMPLE FROM LIGHT. In practice this means putting your sample and request form into a brown envelope. Ultraviolet light will degrade bilirubin at a rate of about 0.005AU (absorption units) per hour. Given that only 0.007AU are required to diagnose subarachnoid haemorrhage, you could easily end up with a false negative CSF result.

The last comment is that use of pneumatic tube delivery may agitate a CSF sample sufficiently to cause haemolysis, relaeasing oxyhaemoglobin and making CSF spectrophotometry more difficult to interpret.

In summary:

• Generous samples
• Paired serum samples
• Protect the sample from light (especially if assaying bilirubin)
• Get the sample to the lab as soon as possible, avoiding pneumatic tube delivery if possible

It seems strange that a lumbar puncture is not more noteworthy.

The humble lumbar puncture is in the background, and the honourable mention goes to the cerebrospinal fluid or neurological diagnosis!

I’ve searched medicalnewstoday.com for news items which mention lumbar puncture.  I managed to find about 19 out of 102,000 (that’s only 0.018%) of all their recorded news items! They included a case of rabies, the articles about the MMR vaccination in autism investigation, research into stem cells and an article on brain cancer. There was one mention of my favourite topic - investigation of acute headache.

Here’s the medicalnewstoday.com news feed, which complies with their terms and conditions….

Lyme-Like Illness To Be Studied In Texas
Tao Lin, D.V.M., and Steven J. Norris, Ph.D., both with the Department of Pathology and Laboratory Medicine at The University of Texas Medical School at Houston, have been named grant recipients of the Norman Hackerman Advanced Research Program (A...

FDA Approves First Generic Divalproex Sodium To Treat Seizures, Migraine Headaches And Bipolar Disorder
The U.S. Food and Drug Administration approved the first generic versions of Depakote delayed-release tablets (divalproex sodium). Depakote is approved by the FDA for the treatment of seizures, bipolar disorder and migraine headaches. "Generic...


Newsfeed display by CaRP

The usual reason for not measuring opening pressure at lumbar puncture is forgetting to include a manometer (pressure guage) in your equipment.

At LP, opening pressure should be measured without exception in the investigation of acute headache. Always remember to measure opening pressure with your patient lying on their side, not sitting upright. The zero point for opening pressure is at the base of the skull, so if you are sitting up and you needle is in the lumbar spine the opening pressure will be at least 30cm (or whatever height the base of the skull is above your needle! If you then add the 10-20cm of intracranial pressure you have a reading in excess of 40cm guaranteed! This is not to be confused with intracranial hypertension, it is a simple but potentially serious error, but it is easy to forget the basics.

When I use Lumbar Puncture to investigate for inflammatory disorders, such as multiple sclerosis, I will deliberately sit the patient up if I don’t have very clear anatomical landmarks, as opening pressure is not crucial in this situation.

However, you need to maintain your skill in Lumbar Puncture with your patient lying on their side, so that if you have to do one for an emergency presentation with headache you have the necesary skill to get the needle inserted with your patient on their side - allowing you to measure opening pressure.

Manometry is the measurement of CSF pressure using a 40cm long plastic gauge. The most common error is not to attemmpt to measure pressure! Measuring pressure is essential in patients presenting with acute headache, as a high or low opeinng pressure may be the only indication that there is a treatable disorder. Idiopathic intracranial hypertension is a disorder of elevated CSF presure, and all other CSF tests will be normal. Similarly, Spontaneous Intracranial Hypotension is a disorder of abnormally low intracranial pressure and all other CSF tests will be normal except for an abnormally low pressure (or CSF is not obtained at all despite good technique - also called a dry tap).

Fact # 4 - always measure opening pressure, it may be the only chance you have to clinch an important diagnosis.

I’ve written my own guideline for doing Lumbar Puncture for my own hospital.  This should reflect best practice.  You can copy this for your own hospital, and adapt it where necessary.  Please remember to quote your source - it is copyrighted, but available for free so long as www.lumbarpuncture.net is acknowledged in your guidance.  Remember that this site contains video links from YouTube so that you are left in no doubt what an LP entails.  Patient information sheet will follow next week.

Here’s the link to the LP Guideline - right click here, then save target as….

22 guage atraumatic needles are now recommended for diagnostic Lumbar Puncture (LP). An atraumatic needle has a rounded tip and is said to avoid cutting the fibres of the dura. If the fibres of the dura are separated without cutting, and the resulting hole is less likely to persist, the risk of causing sustained leak of CSF (cerebrospinal fluid) and low pressure headache is reduced. Some researchers have said that atraumatic needles stil create a tear in the dura, and that it is the local inflammation/healing reaction which reduces the risk of CSF leak.

The issue of needle selection is almost resolved. The most up to date recommendation from the American Academy of Neurology is that 22 guage atraumatic needles should be used, as they are asociated with the lowest risk of an LPs most frequent complication - the post-LP headache. 

The evidence for this comes mostly from anaesthetic literature, but there is little doubt from e.g. good quality studies that an atraumatic needle may reduce post-LP headache from 24% to 12%.

However, not all doctors have ever used atraumatic needles, and the lower risk of post-LP headache should be measured against the risk of a poor patient experience caused by an operator using a needle with which they are not experienced.  A misdirected LP needle can be very painful, particularly if the needle comes into contact with bone (periosteum) or nerve root. This may seem silly i.e. surely all needles are the same?  In actual fact you do get used to the feel of certain needles.  A fine needle is more flexible, whereas you have more control over the direction of a thicker/broader needle. 

There are potentially other methods of preventing post-LP headache.  An interesting way, but not widely accepted, is to insert the needle tangentially i.e. deliberately placing the needle at an angle instead of perpendicular to the surface of the back (see Dr Hatfalvi’s article published in Regional Anesthesia 1995. Tangential insertion also requires training and anatomical knowledge, and image guidance, such as ultrasound, may help less experienced operators.

A preference must surely be for correct use of 22 guage atraumatic needles, but for those of us who are confident with existing cutting or traumatic needles, we will need to justify our continued use of these with evidence of minimal patient discomfort at the actual time of their LP.

The distance from skin to the ligamentum flavum (the tough ligament just befire you reach the epidural space) is between 4 and 7cm in adults. Does anyone know an up-to-date study of this (i.e. if we are getting more obese, is this still true today?)