Lumbar puncture for every negative CT scan?
In their article on subarachnoid haemorrhage (SAH) Al-Shahi and
continue the fallacy that as 2% of SAH is missed on a Computed Tomography
(CT) scan every patient must have a lumbar puncture (LP). Quoting the
sensitivity of CT scanning in this way may lead the unwary into the trap
thinking that about 1 in 50 LPs will be positive after a negative CT (I
probably consent to an LP if there was a 1 in 50 chance of finding a
condition such as a SAH).
However a Bayesian thinker might not be so easily convinced. The
quoted in the article of a sudden headache being due to SAH is 25%, which
can use as the pre-test probability before the CT scan. The quoted
of CT scanning is 98%, giving a Likelihood Ratio for a negative test of
pre-test probability of 25% and a likelihood ratio of 0.02 give a posttest
probability of 0.5%. (I don’t think that I would consent to an LP if there
only a 1 in 200 chance of finding a SAH).
In real practice the increasing ease of access to CT scanning means
are scanning a much lower risk group of patients than in the past. As the
article emphasises, clinical diagnosis of SAH can be very difficult, so CT
are frequently ordered. So the real life pre-CT probability is about 10%.
gives a post-test probability of about 0.15%. (I certainly would not
an LP if there was only a 1 in 650 chance of finding a SAH).
My conclusion is that the dogma that ‘every patient with a “query
SAH” and a
negative CT MUST undergo a lumbar puncture’ is wrong(1). I would suggest
that every patient with a “query SAH” and a negative CT should have a
discussion about a lumbar puncture with a clinician who understands
diagnostic testing so that they can together weigh the risks and benefits,
explain the uncertainty, take the patient’s attitude to risk into
and come to a conclusion about whether further investigation should be
1) Diagnosis of subarachnoid haemorrhage following a negative
tomography for acute headache: A Bayesian analysis. Coats TJ, Loffhagen R
European Journal of Emergency Medicine 2006; 13; 80 – 83.
Competing interests: No competing interests