Article Text

Download PDFPDF

Timing of lumbar puncture in suspected subarachnoid haemorrhage
  1. Simon Carley, Consultant,
  2. Magnus Harrison, Specialist Registrar
  1. Department of Emergency Medicine, Manchester Royal Infirmary, Oxford Road, Manchester M13 9WL, UK;


    A short cut review was carried out to establish how long after onset of headache a lumbar puncture should be carried out to rule out subarachnoid haemorrhage. Altogether 142 papers were found using the reported search, of which one presented the best evidence to answer the clinical question. The author, date and country of publication, patient group studied, study type, relevant outcomes, results and study weaknesses of this best paper are tabulated. A clinical bottom line is stated.

    Statistics from

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Report by Simon Carley, ConsultantChecked by Magnus Harrison, Specialist Registrar

    Clinical scenario

    A 24 year old man presents to the emergency department (ED) with a sudden, severe occipital headache. He collapsed at the time of the initial headache but now feels better. He had computed tomography performed in the ED, which was negative. He was subsequently referred to the medical team who performed a lumbar puncture (LP) one hour after admission (two hours after the initial headache). This was negative and he was permitted home. One week later he re-presents to the ED by ambulance after another collapse. He is GCS 3 on arrival and dies shortly afterwards. Computed tomography and postmortem examination show the cause of death to be subarachnoid haemmorhage. You wonder if the LP was done too early to spot the original bleed.

    Three part question

    In [patients with suspected SAH but a negative CT scan] is [late LP (>12 hours) better than early LP] at [definitivly diagnosing SAH]?

    Search strategy

    Medline 1966–10/04 using the Ovid interface. [(exp subarachnoid hemorrhage OR OR subarachnoid AND (exp cerebrospinal fluid OR spinal OR exp spinal puncture OR lumbar OR AND ( OR tim$.mp)] LIMIT to human, English AND abstracts.

    Search outcome

    Altogether 142 papers were found of which one was relevant to the clinical question (table 1).

    Table 1


    It is common practice to withold LP until 12 hours after the headache onset. This is based on limited evidence from a small number of papers in this review. Most patients in studies of bilirubin biokinetics had positive CT scans. As LP is normally reserved for those patients with a negative CT scan they are arguably a different group. Despite these limitations current laboratory work suggests that bilirubin will remain undetectable until 12 hours after symptom onset. This should remain the current practice. What is not shown from the literature is that any patient who had negative initial findings (on early LP) followed by positive findings (on late LP). Such cases would provide a convincing argument, but none were found.


    In patients with suspected subarachnoid bleeds, LP is not an adequate rule out strategy until 12 hours after the headache onset.

    Report by Simon Carley, ConsultantChecked by Magnus Harrison, Specialist Registrar