Intended for healthcare professionals

Rapid response to:

Clinical Review

Subarachnoid haemorrhage

BMJ 2006; 333 doi: (Published 27 July 2006) Cite this as: BMJ 2006;333:235

Rapid Response:

Re : Subarachnoid haemorrhage : lumbar puncture for every negative scan

Professor Coats makes an interesting argument for not performing
lumbar puncture on every patient whose CT scan is negative for
subarachnoid haemorrhage (SAH). It is, I fear, a rather flawed agrument.
Whilst CT scan has a published sensitivity of 98% for the detection of
SAH, the performance of diagnostic imaging outside a clinical trial or
centre of interest may be significantly worse than this. The results of
the PIOPED II study have published a sensitivity of CT pulmonary
angiography of 83% [1], much lower than previously published; it seems
illogical that CT for possible SAH would perform differently.

The increasing availability of a test should not mean that it is
applied less discriminately, as Professor Coats implies. All uses of
ionizing radiation must be justified under radiation protection
legislation, which is part of criminal law. If he believes that CT is
indeed being used less discriminately then this would seem to more an
argument about the experience and training of those making the request
rather than what to do with the test result. Further, his pre-test
probability of SAH in the real world of 10% seems an arbitrary figure. If
the real world performance of CT is a sensitivity of 90%, an equally as
arbitrary but no less realistic figure, then the post-test probability
could be as high as 1%. Would he perform a lumbar puncture in such

What is clear, is the outcome of patients in whom a SAH is missed.
Untreated SAH has a 40% rebleed rate at one month [2], with a large
proportion of patients either dying or left significantly impaired. With
the treatment now available for aneurysms and arteriovenous malformations,
not performing a lumbar puncture will miss potentially treatable disease
at as stage where outcome is liekly to be good. Professor Coats' takes no
account of this.

Whilst Bayesian analysis may be helpful in considering circumstances
where the literature is not clear, it cannot be said that CT scanning and
SAH is such a case. If a doctor is sufficiently concerned of the risk of
SAH to expose a patient to ionizing radiation for a head CT, then they
should be prepared to perform a lumbar puncture should this be negative.
This may be a dogmatic approach but it is at least evidence-based.

1. Stein PD, Fowler SE, Goodman LR, et al. Multidetector computed
tomography for acute pulmonary embolism. N Engl J Med 2006;354(22):2317-

2. Locksley HB. Natural history of subarachnoid hemorrhage, intracranial
aneurysms and arteriovenous malformations. Based on 6368 cases in the
cooperative study. J Neurosurg 1966;25(2):219-39.

Competing interests:
EMA is occasionally asked for CT head scans in the early hours of the morning

Competing interests: No competing interests

19 August 2006
Ewan M Anderson
Specialist Registrar Radiology