INVESTIGATIONS

Lumbar punctures and cerebrospinal fluid analysisJustin Pearson

Geraint Fuller

AbstractLumbar puncture is an essential neurological investigation. The anatomy

of the spinal canal allows cerebrospinal fluid (CSF) to be sampled from

the lumbar region. We describe how, with adequate preparation and cor-

rect positioning of the patient, the procedure can be performed quickly

and safely. The range of investigations that can be performed on the CSF

and their interpretation will be discussed.

Keywords cerebrospinal fluid; diagnostic techniques; lumbar puncture

Lumbar puncture and analysis of cerebrospinal fluid (CSF) are a routine part of clinical practice both in the setting of acute medi-cal admissions and in the routine investigation of neurological diseases. It is a safe procedure as long as you are aware of the risks and contraindications (see below). In the majority of cases it is performed to aid the diagnosis of central nervous system infections, subarachnoid haemorrhage, multiple sclerosis and malignancy.

Anatomy and physiology

Three layers of meninges cover the brain and spinal cord: the dura, arachnoid and pia. The CSF is contained within the sub-arachnoid space. The CSF is mostly produced by the choroid plexuses in the ventricles and flows from the lateral ventricles through the third and fourth ventricles into the subarachnoid cis-tern around the medulla and on over the surface of the brain and spinal cord where it is reabsorbed by the arachnoid villi.

Within the spinal canal, the spinal cord ends at approximately the level of the first lumbar vertebra. Below this level the nerve roots of the cauda equina float in the CSF. A needle can therefore be passed between the spinous processes of two adjacent lower lumbar vertebrae to enter the spinal canal by puncturing the dura (Figure 1).

Justin Pearson MRCP BSc is a Neurology Specialist Registrar in the

Severn Deanery, UK. He qualified from University College London.

Competing interests: none declared.

Geraint Fuller MA MRCP is Consultant Neurologist at Gloucestershire

Royal Hospital, Gloucester, UK. He qualified from the University of

Cambridge and St Mary’s Hospital, London. His particular interests are

training and education. Competing interests: none declared.

MEDICINE 36:10 562

Procedure

Prior to performing a lumbar puncture (LP), care should be taken to ensure that it is indicated and that there are no contraindications (see Tables 1 and 2). The most important contraindication to LP is the presence of an intracranial mass lesion. In this situation, an LP may result in cerebral herniation and death. Thus, for patients with focal neurological signs, seizure, impaired consciousness or papill-oedema, brain imaging should be performed prior to an LP.

ConsentInformed consent should be obtained. Patients should be warned about the potential complications of headache and backache. The headaches are characteristic ‘low-pressure’ headaches: they are generalized and markedly postural, worse on sitting and eas-ing on lying flat. Most are self-limiting, but can require treatment with autologous blood patch. The frequency of post-lumbar

Spinal anatomy and needle positioning

Diagrammatic illustration of the spinal anatomy that allows lumbar puncture, seen in axial and sagittal planes.

Figure 1

© 2008 Elsevier Ltd. All rights reserved.

INVESTIGATIONS

puncture headache has been reduced from 32% to 6% by using a 22G pencil-point needle compared to the traditional 20G bevelled needle.1 More serious, though very rare, complications include epidural haematoma or infection.

PreparationPreparation prior to performing a lumbar puncture is essential. A well-lit room with a firm, height-adjustable couch and an assistant should be arranged. The equipment required should be prepared on a sterile field. This will include an appropriate lumbar puncture needle. As detailed above, pencil-point needles are preferred to the traditional bevelled needle. However, when the lumbar puncture is being performed to reduce CSF pressure, for example in idio-pathic intracranial hypertension, a bevelled needle can be used.

Correct patient positioning is the key to successful lumbar puncture. The patient should be asked to lie in the left lateral position (for the right-handed operator) with their back along the edge of the couch. They should be asked to adopt the foetal position with their neck, hips and knees flexed. A single pillow should be placed under the patient’s head and one between their knees. Care should be taken to ensure that their hips lie vertically above each other and likewise the shoulders.

The operator should then examine the patient’s back to identify the anatomical landmarks (Figure 2). First, locate the iliac crest and then palpate the spinous processes. The level vertically below the iliac crest should be between L3 and L4. This should be identified by careful palpation as visible surface landmarks can be misleading.

TechniqueA sufficiently large area of surrounding skin should then be sterilized to maintain a sterile field. Lidocaine (2%) can then be

Contraindications

• Symptoms and signs of focal intracranial neurological disease

or raised intracranial pressure including confusion (unless it

has been determined by neuroimaging that it is safe)

• Neuroimaging evidence of obstruction to cerebrospinal fluid

flow

• Coagulopathy including thrombocytopenia and prolonged

clotting time (INR >1.3)

• Signs of local infection at the site of skin puncture

Table 1

Indications

Cerebrospinal fluid examination is often useful in the diagnosis of:

• Subarachnoid haemorrhage

• Meningitis/encephalitis

• Inflammatory neurological disorders including multiple

sclerosis and Guillain-Barré syndrome

• Carcinomatous or lymphomatous meningitis

• Prion disease

• Idiopathic intracranial hypertension

Table 2

MEDICINE 36:10 563

infiltrated into the skin overlying the intervertebral space as far as the intervertebral ligament. While the anaesthetic takes effect, put the manometer together.

The needle is then inserted in the midline pointing towards the umbilicus. Once beyond the subcutaneous fat a steady level of resistance should be felt as the needle passes through the supraspinous and interspinous ligaments. If using a bullet or pen-cil point needle you will start with the introducer till you pass through the ligaments, and then pass the blunt needle through the introducer. An additional brief increase in resistance may then be felt as the needle passes through the dura before a feeling of give as the needle passes into the subarachnoid space. At this point the stylet should be removed and there should be backflow of CSF.

A three-way tap and manometer is attached and the CSF pres-sure measured. At least three samples of 2–5 ml CSF should then be taken as well as a sample in a fluoride bottle for glucose. Additional volumes of CSF may be required depending on the number of tests being requested (e.g. biochemistry, bacteriology, virology – Table 3).

Technical difficultiesIf CSF is not obtained, the stylet can be reinserted and the needle advanced further before checking for CSF again. If the needle cannot be advanced further, then bone has been encountered and the needle should be removed, the angle of insertion adjusted and the process repeated (Figure 1). If CSF is still not obtained, the process can be repeated at a lower level or with the patient seated (although this method does not allow measurement of opening pressure).

Multiple failed attempts should be avoided as this may lead to increased back pain and trauma. There should be a low thresh-old for seeking a more experienced operator or arranging for the

Anatomical landmarks

Illustration of patient position and anatomical landmarks shown from the operator’s point of view.

L3 L4

Iliaccrest

Figure 2

© 2008 Elsevier Ltd. All rights reserved.

INVESTIGATIONS

procedure to be performed under fluoroscopy. Very rarely CSF will not be obtained because of an infiltrating intraspinal lesion.

Interpretation

Table 4 summarizes the findings on CSF analysis in both normal subjects and common neurological diseases.

Interpretation of CSF findings in meningitides is very impor-tant. The cell count and differential need to be considered along

Tests performed on cerebrospinal fluid (CSF)

Routine CSF examinination

• Opening pressure

• Microbiology – microscopy and cell count

• Biochemistry – protein and glucose

CSF tests performed only in appropriate circumstances

• Microbiology – culture for bacteria or tuberculosis (TB),

polymerase chain reaction (PCR) for viruses or TB, special

stains for fungi or TB, cryptococcal antigen, protein 14-3-3

for Creutzfeld–Jakob disease

• Biochemistry – oligoclonal bands, spectroscopy

• Cytology – microscopy, cytospin

• Haematology – immunophenotyping and fluorescence-

activated cell sorting (FACS) analysis

Table 3

MEDICINE 36:10 56

with the CSF protein and glucose results – see Table 4. The CSF glucose level needs to be compared to a matched blood glucose. It is worth making a point of writing all these results together in the notes to aid interpretation. CSF pressure results can also provide useful information − a high CSF pressure with normal CSF may indicate cerebral venous thrombosis.

Special situations

Subarachnoid haemorrhage (SAH)During SAH red cells enter the CSF and their breakdown releases oxyhaemoglobin into the CSF. This oxyhaemoglobin is then con-verted into bilirubin. This conversion process can occur only in vivo. Therefore, detection of high levels of bilirubin in the CSF is the most diagnostic finding for SAH but is not 100% specific. CSF spectroscopy can detect the blood breakdown products of SAH up to 2 weeks after the event.2 An isolated finding of oxyhaemo-globin is difficult to interpret as it may represent haemolysis of red cells introduced into the CSF at the time of the lumbar punc-ture. This test should be performed on the last of the samples taken. The sample should be protected from the light to prevent further breakdown of the red cells.

MalignancyA sample of CSF can also be spun down to look for abnormal or malignant cells in the investigation of central nervous system malignancy.3 Any abnormal cells can then be further analysed by fluorescence-activated cell scanning (FACS) and immuno-phenotyping, which is particularly used in the diagnosis of CNS

Typical cerebrospinal fluid findings

Leucocytes Protein Glucose Additional tests

Normal <5 cells/mm3 0.15–0.45 g/L

(age dependent)

>2/3 of serum glucose Opening pressure 5–20

cm H2O

Bacterial meningitis High to very high cell

count, often >1000 PMNs/

mm3 (lymphocytosis when

partially treated or listeria)

Elevated, usually >1.0 g/L Very low, usually

<40% of serum level

Gram stain and culture

PCR for some bacteria

Tuberculous meningitis Modest lymphocytosis

(or PMNs early in the

disease)

Markedly elevated >1.0 g/L Low <40% of serum

level

Elevated opening presssure

Microscopy for AFBs

TB culture and PCR

Viral meningitis/

encephalitis

Modest lymphocytosis

usually less than 200

cells/mm3

Normal or mildly elevated Normal or mildly low Viral PCR

Fungal meningitis Lymphocytosis <50

cells/mm3

Normal then elevated Normal or mildly low Indian ink stain

Cryptococcal antigen

Multiple sclerosis <10 cells/mm3 Normal Normal Positive oligoclonal bands

in 90% of cases

Other inflammatory CNS

diseases

Normal or lymphocytosis Normal Normal Oligoclonal bands often

present

Inflammatory PNS

disease e.g. AIDP/CIDP

Normal Increased, may be normal

initially

Normal None

PMN, polymorphonuclear neutrophil; PCR, polymerase chain reaction; AFB, acid fast bacilli; CNS; central nerrous system; PNS, peripheral nervous system; TB, tuberculosis; AIDP, acute inflammatory demyelinating polyneuropathy; CIDP, chronic inflammatory demyelinating polyneuropathy.

Table 4

4 © 2008 Elsevier Ltd. All rights reserved.

INVESTIGATIONS

lymphoma.4 Repeated lumbar punctures are often required to identify malignant cells.

DementiaMeasurement of the brain-specific protein 14-3-3 in CSF is used in the diagnosis of rapidly progressive dementia.5 In the appro-priate clinical scenario, an elevated level is very suggestive of Creutzfeld–Jakob disease (CJD), but is not entirely specific as it can be elevated in other rapidly progressive dementias and fol-lowing prolonged seizures. Other potential CSF biomarkers for dementia are also being studied including CSF tau and amyloid, but these are not currently used in routine clinical practice. ◆

ReFeRenCeS

1 Kleyweg RP, Hertzberger LI, Carbaat PA. Significant reduction in

post-lumbar puncture headache using an atraumatic needle.

A double-blind, controlled clinical trial. Cephalalgia 1998; 18: 635–37.

2 UK National External Quality Assessment Scheme for

Immunochemistry Working Group. National guidelines for analysis

of cerebrospinal fluid for bilirubin in suspected subarachnoid

haemorrhage. Ann Clin Biochem 2003; 40: 48–88.

MEDICINE 36:10 565

3 Gleissner B, Chamberlain MC. Neoplastic meningitis. Lancet Neurol

2006; 5: 443–52.

4 Urbanits S, Griesmacher A, Hopfinger G, et al. FACS analysis – a new

and accurate tool in the diagnosis of lymphoma in the cerebrospinal

fluid. Clin Chim Acta 2002; 317: 101–07.

5 Van Everbroeck B, Boons J, Cras P. Cerebrospinal fluid biomarkers

in Creutzfeldt-Jakob disease. Clin Neurol Neurosurg 2005; 107:

355–60.

Practice points

• CSF analysis is an essential part of investigation of

neurological disease

• Lumbar puncture is a safe routine procedure

• Adequate preparation and correct patient positioning are

essential for successful completion of a lumbar puncture

• Accurate diagnosis depends on requesting the appropriate

CSF tests and interpreting the results in the clinical context

© 2008 Elsevier Ltd. All rights reserved.