01Purpose and principlesWhat the assessment is for and the core concepts behind it.
Lumbar puncture samples the subarachnoid compartment and can measure pressure, but its diagnostic value depends on preparation. The clinician should decide why CSF is needed, whether the result will change treatment, whether immediate therapy comes first and which samples must be paired with blood. Calling the laboratory or specialist team before a difficult or high-stakes puncture prevents inadequate volumes, wrong containers and lost molecular, cytological or spectrophotometric testing.
Contraindications are contextual. Local infection over the puncture site, severe cardiorespiratory instability, uncorrected major bleeding risk and concern for a dangerous intracranial pressure gradient can make immediate LP unsafe. Focal deficit, markedly impaired consciousness, recent seizure, papilloedema or immunocompromised state may prompt urgent imaging and senior review depending on the syndrome. A normal CT cannot exclude every pressure gradient, and empirical treatment for CNS infection must continue while the procedure is deferred.
Standard adult technique usually uses L3/4 or L4/5, below the conus, with the intercristal line as a surface guide. Position, sterile preparation, local anaesthesia, atraumatic needle orientation and patient coaching influence success and post-puncture headache. Ultrasound can identify midline and depth in difficult anatomy; fluoroscopic puncture is a planned escalation rather than repeated traumatic bedside attempts.
Interpretation integrates opening pressure, appearance, cell count and differential, protein, glucose relationship, microscopy, culture and molecular testing. Antibiotics can sterilise cultures while PCR remains informative; early infection can have modest cells; a traumatic tap can raise cells and protein; oligoclonal bands require paired serum; malignant cytology depends on volume and rapid processing. No isolated CSF value should override a severely unwell clinical phenotype.
Key points
- Define the question before puncture: infection, inflammation, subarachnoid haemorrhage, malignant infiltration, pressure disorder and therapeutic drainage require different timing, tubes, paired blood tests and laboratory handling.
- Obtain consent covering purpose, alternatives, discomfort, bleeding, infection, post-dural-puncture headache, failure and the rare risk of neurological deterioration when contraindications are missed.
- Check observations, conscious level, focal signs, optic discs where competent, anticoagulants, antiplatelets, platelet count and coagulation context, spinal anatomy, local infection and cardiorespiratory ability to tolerate positioning.
- Brain imaging before LP is selective, not routine. Shock, airway compromise or strong mass-effect risk changes the order; imaging itself does not make puncture safe if clinical contraindications persist.
- Use an aseptic technique and an atraumatic small-gauge needle when feasible. Position the patient laterally for a valid opening pressure, relaxed with legs no more flexed than needed after the interspace is entered.
- Measure opening pressure at equilibrium with the manometer zeroed at the needle, recording position, sedation, straining and technical limitations. A number obtained seated is not interchangeable with standard lateral measurement.
- Plan sample allocation before starting and label tubes in sequence. Typical studies include paired glucose, protein, cells and differential, Gram stain and culture, with PCR, cytology, oligoclonal bands or specialist tests added for the actual syndrome.
- CSF glucose must be interpreted against a near-simultaneous blood glucose. A low CSF-to-blood relationship can support bacterial, tuberculous, fungal or malignant disease but is not exclusive to one cause.
- Neutrophilic pleocytosis, high protein and low glucose suggests bacterial meningitis, yet early viral infection, partially treated disease and immunosuppression can produce atypical patterns; microbiology and clinical severity govern action.
- For suspected aneurysmal subarachnoid haemorrhage after negative CT, follow NICE NG228: CT timing matters and, when LP is indicated, allow at least twelve hours from symptom onset before spectrophotometric bilirubin analysis.
- A falling red-cell count across tubes does not reliably distinguish a traumatic tap from subarachnoid bleeding. Use bilirubin spectrophotometry, CT timing and the whole clinical pathway rather than visual inspection alone.
- Communicate urgent samples directly to the laboratory, record antimicrobial timing and transport requirements, review every final culture and PCR result and arrange a plan for unresolved or incidental abnormalities.
02Indications, selection and cautionsWhen it is useful, when urgency changes and important limitations.
Raised opening pressure, turbid fluid, neutrophilic pleocytosis, increased protein and reduced CSF glucose relative to blood supports bacterial meningitis, but prior antibiotics, early disease and immunosuppression can alter every component.
Lymphocytic pleocytosis with moderately increased protein and usually preserved glucose supports viral infection. Early viral disease may be neutrophilic, and HSV encephalitis can have red cells, so PCR and phenotype are essential.
Raised pressure, lymphocytic cells, high protein and low glucose can suggest tuberculosis or fungal disease. Yield depends on volume, repeated sampling and specialist microbiology, and empirical decisions may precede confirmation.
CSF bilirubin detected by validated spectrophotometry at the correct interval can support subarachnoid haemorrhage after appropriate CT assessment. Oxyhaemoglobin alone is vulnerable to artefact from traumatic sampling.
CSF-restricted oligoclonal bands or an increased IgG relationship can support intrathecal inflammation in the correct clinical and MRI context. They are not specific to multiple sclerosis and require paired serum.
Raised protein with few cells can support Guillain–Barré syndrome after the first week, but early CSF may be normal and a substantial pleocytosis should prompt infection, malignancy or another diagnosis.
A reproducibly raised lateral opening pressure with an otherwise appropriate CSF profile can support intracranial hypertension, but obesity, Valsalva, pain, sedation, flexion and positioning can falsely elevate the reading.
03Method and interpretationA systematic approach to the test and its findings.
Read from the initial assessment onwards. Tests may run in parallel in urgent care; first-line, preferred, confirmatory, definitive and gold-standard labels appear only when the chapter explicitly states them.
- 01
Opening pressureFirst step - Why
- Measure subarachnoid pressure for suspected intracranial hypertension, infection and selected other syndromes.
- Interpretation and limitations
- Record lateral position and confounders and wait for equilibrium. Thresholds require age and clinical context; a technically difficult isolated value should not establish idiopathic intracranial hypertension.
- 02
Cell count, differential and red-cell count - Why
- Identify inflammation, its dominant cell type and blood contamination or haemorrhage.
- Interpretation and limitations
- Cells can degrade with delay, so process promptly. Formulae that 'correct' white cells for a traumatic tap are approximate and should not dismiss meningitis in a compatible unwell patient.
- 03
CSF protein - Why
- Assess blood–CSF barrier dysfunction, inflammation, nerve-root disease and impaired CSF flow.
- Interpretation and limitations
- Protein rises with age, blood contamination, infection, tumour, spinal block and inflammatory neuropathy. Magnitude and accompanying cells are more useful than a binary high result.
- 04
Paired CSF and blood glucose - Why
- Evaluate glucose consumption and transport in infectious, inflammatory and malignant meningitis.
- Interpretation and limitations
- Obtain blood glucose close to puncture and interpret a ratio or paired relationship. Systemic hyperglycaemia and timing affect the values; low CSF glucose is concerning but not uniquely bacterial.
- 05
Microscopy, culture and pathogen PCR - Why
- Identify bacterial, viral and selected fungal or mycobacterial causes and enable antimicrobial refinement.
- Interpretation and limitations
- Prior treatment reduces culture yield, while PCR panels are limited to included targets and timing. Discuss additional volumes and specialist tests with microbiology before puncture when possible.
- 06
Bilirubin spectrophotometry - Why
- Assess suspected subarachnoid haemorrhage after negative CT when NICE timing indicates lumbar puncture.
- Interpretation and limitations
- Take the final protected sample at least twelve hours after onset when the test is indicated, follow the laboratory's handling instructions and interpret bilirubin with serum and CT findings.
- 07
Paired oligoclonal bands and cytology or flow cytometry - Why
- Investigate intrathecal inflammation or leptomeningeal malignancy in selected syndromes.
- Interpretation and limitations
- Oligoclonal interpretation requires serum collected at the same episode. Cytological yield improves with adequate volume, rapid transport and sometimes repeat sampling guided by haematology or oncology.
04Clinical next stepsHow the result changes management or prompts escalation.
01Suspected CNS infectionTreat before unsafe delayFirst stepFever, headache, meningism, altered behaviour, reduced consciousness, focal deficit, rash or seizure creates concern for meningitis or encephalitis.+
- 1Perform ABCDE, obtain blood cultures and other immediate samples and start guideline-concordant empirical antibiotics and, when indicated, aciclovir and adjunctive therapy without waiting for imaging or LP if delay would result.
- 2Assess cardiorespiratory stability, conscious level, focal signs, papilloedema risk, recent seizure and haemostasis to decide whether puncture can occur now or needs CT and specialist review first.
- 3Send planned CSF urgently with paired glucose, document pre-treatment timing and refine isolation, antimicrobial and public-health actions with microbiology, infection and national guidance.
02Suspected subarachnoid haemorrhageCoordinate CT and bilirubin timingThunderclap headache or compatible collapse, meningism or focal syndrome with no haemorrhage shown on initial CT.+
- 1Confirm CT quality and exact interval from headache onset and discuss persistent suspicion with the emergency, stroke or neurosurgical team rather than ordering LP automatically.
- 2If CT occurred within six hours and was reported by a radiologist as negative, follow NICE advice not to routinely offer LP; if CT was after six hours, consider LP after specialist discussion.
- 3When puncture is chosen, wait at least twelve hours from onset, send the designated protected sample for bilirubin spectrophotometry and route positive or unresolved cases urgently.
03Planned diagnostic LPMaximise safety and yieldInflammatory, pressure, neuropathic or malignant disease is suspected without current cardiorespiratory or mass-effect emergency.+
- 1Agree the clinical question and tests with neurology, microbiology or haematology as needed, review medicines and haemostasis, obtain consent and gather atraumatic equipment, labels and paired blood tubes.
- 2Position laterally when pressure is required, use aseptic technique and local anaesthesia, record opening pressure and collect sufficient sequential tubes with minimal repeated attempts.
- 3EscalationProvide post-procedure advice, review urgent and final results, contact the patient or team with the interpretation and arrange escalation for headache, infection, bleeding or new neurological symptoms.
05Procedure and medicine safetyRelevant preparation, treatment and contraindications.
Lidocaine 1% local anaesthetic
Infiltrate the skin and deeper tract incrementally after aspiration, using the smallest effective volume and remaining below the local maximum dose, commonly 3 mg/kg without adrenaline.Confirm allergy, avoid intravascular injection, calculate all lidocaine exposure and monitor for neurological or cardiovascular toxicity; an effective skin wheal alone may not anaesthetise the deeper periosteal tissues.
06Risks, monitoring and follow-upComplications, safety checks and further assessment.
- Observe immediate neurological state, comfort and the puncture site according to clinical setting, with longer monitoring when physiology or bleeding risk was concerning.
- Track post-dural-puncture headache by postural nature, severity, hydration and neurological features; atypical, non-postural or progressive headache needs reassessment for another complication.
- Review preliminary microscopy and Gram stain promptly, then actively follow culture, PCR, oligoclonal, cytology and specialist results rather than assuming discharge closes the episode.
- Document opening-pressure position, needle type, level, attempts, sample order, antimicrobial timing and any traumatic or incomplete collection for later interpretation.
- After a deferred puncture, record the reason, empirical treatment and explicit criteria and ownership for reconsidering CSF sampling.
- Safety-net fever, wound discharge, severe persistent headache, new weakness, sphincter change, confusion or reduced consciousness with a route for urgent review.
07Special situationsVariants, exceptions and circumstances that change the usual approach.
Imaging is not clearance
A normal CT does not guarantee LP safety because pressure gradients and clinical instability are not excluded by one image. The bedside syndrome and prerequisites remain decisive.
Atraumatic needles matter
Pencil-point atraumatic needles reduce post-dural-puncture headache and return-to-hospital burden. Use an introducer and familiar technique rather than defaulting to a cutting needle.
Pressure is posture-dependent
Seated measurement, breath-holding, pain and excessive hip flexion can raise readings. Standard lateral positioning and documentation make the number clinically interpretable.
Tube clearing is unreliable
Red cells may fall in both traumatic tap and genuine haemorrhage. Modern UK subarachnoid pathways rely on CT timing and laboratory bilirubin analysis, not a three-tube visual rule.
Normal early CSF can occur
Early encephalitis, meningitis and Guillain–Barré syndrome may have non-diagnostic CSF. A strong evolving phenotype deserves treatment, repeat assessment and specialist-guided retesting.
Volume should fit purpose
Mycobacterial culture, cytology and multiple specialist assays may need more fluid than routine chemistry. Pre-procedure planning avoids a second puncture and protects essential first-line tests.
08Common pitfallsFrequent interpretation and management errors.
- 01
Delaying antibiotics in suspected bacterial meningitis until imaging and lumbar puncture are complete.
- 02
Requesting CT routinely before every puncture and mistaking a negative scan for procedural clearance.
- 03
Measuring an opening pressure in the seated position and comparing it with lateral reference values.
- 04
Sending CSF glucose without a paired blood glucose collected near the same time.
- 05
Correcting a traumatic white-cell count with a formula and dismissing a compatible CNS infection.
- 06
Using red-cell decline between tubes to exclude subarachnoid haemorrhage.
- 07
Failing to arrange ownership for delayed culture, PCR, oligoclonal or cytology results.