• 
• 

Three Part Question

In [patients presenting with sudden onset headache] who [walk into the department aged less than 55 years] can [lumbar puncture be avoided]

Clinical Scenario

A 47-year-old woman self-presents to the emergency department complaining of a sudden-onset headache associated with vomiting, which had developed suddenly. Her only medical history is migraine with aura diagnosed and treated by a neurology clinic; but this felt different. Triage notes show that she is apyrexial and routine observations are within normal parameters (GCS E4 V5 M6). No neurological signs are present but she appears incredibly uncomfortable, in the absence of true photophobia. A CT scan is done and no abnormality is identified. The patient feels reassured and is keen to get home, where her husband would be able to keep an eye on her. This seems reasonable. However, you wonder whether or not there is evidence for any circumstances where not progressing to lumbar puncture +/- admission would be supported, despite the current consensus opinion that it is required for the added confidence when combined with CT, in excluding sub-arachnoid haemorrhage (SAH). Your thought is based on the fact that in your experience LP procedures on non-ambulance arrivals have not yielded positive results.

Search Strategy

Medline In-process & other non-indexed citations, EMBASE and Ovid MEDLINE 1948 to March Week 3 2011 using the Athens interface. In addition Cochrane Central Register of Controlled Trials and Cochrane Database of Systematic Reviews 1st Quarter 2011.

Search Strategy: [(headache.m_titl.) OR (head pain.m_titl.) OR (migraine.m_titl.)] AND [(ambulatory.m_titl.) OR (walking.m_titl.) OR (self presenting.m_titl.) OR self referral.m_titl.) OR (minor.m_titl.)] AND [(A & E.m_titl.) OR (emergency.m_titl.) OR (emergency department.m_titl.) OR (accident and emergency.m_titl.) OR (casualty.m_titl.)] AND [(SAH.m_titl.) OR (LP.m_titl.) OR (intracranial bleed.m_titl.) OR (subarachnoid bleed.m_titl.) OR (subarachnoid haemorrhage.m_titl.) OR (brain haemorrhage.m_titl.) OR (subarachnoid hemorrhage.m_titl.) OR (brain hemorrhage.m_titl.) OR (intracranial hemorrhage.m_titl.)

Search Outcome

50 papers were identified in MEDLINE with no further papers identified by an EMBASE search or through the Cochrane Central Register of Controlled Trials. None of the papers looked at or identified populations, or provided data on subgroups, which consistently had negative CT head followed by negative LP. To our knowledge, this analysis has not been attempted retrospectively either.

Relevant Paper(s)

Author, date and country	Patient group	Study type (level of evidence)	Outcomes	Key results	Study Weaknesses

Comment(s)

The incidence of SAH in patients presenting to the ED with headache is 1%. When a typical history is present (severe, sudden-onset headache) and there are no neurological findings, 10% have SAH.1 It is a serious condition that frequently leads to neurological impairment and death2, a missed diagnosis can be devastating for the patient as the risk of re-bleeding is high if the initial bleed is missed and it is a condition for which treatment is possible.

Third generation CT scanners are 90-98% sensitive for SAH within the first 24 hours1, 3 and sensitivity decreases over time as blood is cleared from the CSF because its high density on CT is a function of the haemoglobin concentration. Therefore a non-discriminatory CT scan result in this context should be followed by lumbar puncture (LP) to ensure SAH is not missed. However, LP is not without risks. It is associated with a high rate of both postural puncture headache (up to 30%) and back pain (35%). Occasionally, post LP headaches are unremitting with conservative treatment. Other reported risks include iatrogenic meningitis, nerve palsies, and epidural and subdural haematoma. ‘Traumatic taps’ cause further difficulty in interpretation of results4. Patients undergoing LP are likely to be admitted to hospital and occupy acute medical beds whilst this is undertaken.

We recognize that CT alone is not sufficient for exclusion of SAH. As long as CT remains less than 100% sensitive for SAH, it will remain impossible to randomize patients to an LP and a control group experimentally. Once a decision to undertake head CT has been made, we cannot recommend from review of the identified literature, an accurate clinical risk stratification method for reducing the number of ‘negative’ LP investigations following a CT that doesn’t show evidence of blood.

Clinical Bottom Line

There is no reliable method for clinical risk stratification and investigation planning when SAH is suspected. All patients presenting with acute severe onset headache should have urgent CT if SAH is being considered within the differential diagnosis. If this is negative then LP should be performed with appropriate CSF analysis, irrespective of the patient’s method of arrival, demographics or clinical parameters at the time of presentation.

References

1. Edlow JA. Diagnosis of subarachnoid haemorrhage in the Emergency Department Emerg Med Clin N Am. 2003; 21:73-87.
2. Vemeulen MJ, Shull MJ. Missed diagnosis of subarachnoid haemorrhage in the Emergency Department. Stroke. 2007; 38(4):1216-2
3. Carley S, Wallmann P. Does a normal CT scan rule out a subarachnoid haemorrhage?BestBET Emergency Medicine Journal, 2001 Jul; 18(4): 271-3.
4. Byyny RL, Mower WR, Shum N et al. Sensitivity of non-contrast cranial computed tomography for the Emergency Department diagnosis of subarachnoid haemorrhage. Annals of Emergency Medicine, June 2008; 51(6): 697-703.