Spectrophotometry of cerebrospinal fluid in suspected subarachnoid haemorrhageBMJ 2005; 330 doi: http://dx.doi.org/10.1136/bmj.330.7483.138 (Published 13 January 2005) Cite this as: BMJ 2005;330:138
- Anne Cruickshank, consultant chemical pathologist ()1,
- Robert Beetham, consultant clinical scientist2,
- Ian Holbrook, consultant clinical scientist3,
- Ian Watson, consultant clinical scientist4,
- Philip Wenham, consultant clinical scientist5,
- Geoffrey Keir, consultant clinical scientist6,
- Peter White, deputy director UK NEQAS for immunology and immunochemistry7,
- William Egner, director7
- 1 Department of Biochemistry, Southern General Hospital, Glasgow G51 4TF,
- 2 Frenchay Hospital, Bristol,
- 3 York Hospital, York,
- 4 Walton Centre for Neurology and Neurosurgery, Liverpool,
- 5 Victoria Hospital, Kirckaldy,
- 6 Department of Neuroimmunology, National Hospital for Neurology and Neurosurgery, London,
- 7 Department of Immunology, Northern General Hospital, Sheffield
- Correspondence to: A Cruickshank
- Accepted 25 September 2004
To underline the importance of correctly examining cerebrospinal fluid, we describe two patients with subarachnoid haemorrhage for whom cranial computerised tomography was normal and on whom lumbar puncture was subsequently done. Both had clear and colourless cerebrospinal fluid, but subsequent spectrophotometry showed bilirubin. Cerebral angiography then showed a ruptured aneurysm in each patient.
Case 1A 28 year old woman presented with nausea and diplopia. She had global headache which had awoken her 14 days before. This was associated with neck pain and stiffness, vomiting for two days, and photophobia for five to six days. On examination, severe bilateral papilloedema, and signs of a VIth nerve palsy were present. Cranial computerised tomography showed no abnormality. Lumbar puncture was done. Cerebrospinal fluid contained 79 red cells and five white cells per cubic millimetre. The supernatant was clear and colourless on visual inspection when held up to light. Nevertheless, spectrophotometry detected bilirubin (0.021 absorbance units). Subsequent cerebral angiography showed a ruptured right middle cerebral artery aneurysm.
Case 2A 47 year old woman presented with a history of headache which had begun 10 days before. She reported the start as a few minutes of severe pain shooting up her neck to her head. The pain was associated with some vomiting and neck stiffness but no photophobia. Examination was unremarkable. Cranial computerised tomography showed no abnormality. Lumbar puncture was done on the eleventh day after the onset of headache. Cerebrospinal fluid contained 125 red cells and six white cells per cubic millimetre. The supernatant was clear and colourless on visual inspection when held up to light. Despite this, spectrophotometry of cerebrospinal fluid detected bilirubin (0.009 absorbance units). Subsequent angiography showed a ruptured left posterior cerebral artery aneurysm.
Computerised tomography can miss some cases of subarachnoid haemorrhage. Diagnostic sensitivity has been reported to decrease from 92% on the day of the bleed to 58% on day five.1 Because of this, it has been recommended that where subarachnoid haemorrhage is suspected but computerised tomography is normal, lumbar puncture should be done to exclude the presence of haem pigments (bilirubin and oxyhaemoglobin) in cerebrospinal fluid.2 After subarachnoid haemorrhage, haemolysis of subarachnoid erythrocytes releases haemoglobin, which is converted to bilirubin. Bilirubin concentration reaches a maximum at about 48 hours and may last for two to four weeks after extensive bleeding.3
In the two cases described here, spectrophotometry found increased bilirubin in cerebrospinal fluid providing evidence that subarachnoid haemorrhage had occurred; in each case the appearance of cerebrospinal fluid was normal. Without spectrophotometry the diagnosis may have been missed. Recent published guidelines have emphasised the poor sensitivity of visual inspection of cerebrospinal fluid for xanthochromia (yellow or reddish discolouration), and the importance of spectrophotometry.4 Many laboratories, however, continue to report the absence of haem pigments based on visual inspection.5 Spectrophotometry should be used to confirm the absence of haem pigments in cerebrospinal fluid. After lumbar puncture, the least blood stained sample of cerebrospinal fluid should be protected from light and sent immediately to the laboratory for spectrophotometry.
Spectrophotometry should be used to confirm the absence of haem pigments in cerebrospinal fluid
Contributors All authors had the idea. Details of each case were provided by RB. The paper was written by AC and RB. AC is guarantor.
Competing interests None declared.
Ethical approval Not needed.