M J Wood1, G Dimeski, A M Nowitzke. 1. Department of NeurosurgeryPrincess Alexandria Hospital, Brisbane, QLD, Australia.
Abstract
AIMS: To assess the sensitivity and specificity of CSF spectrophotometry for the detection of xanthochromia in patients with clinical symptoms suggestive of subarachnoid haemorrhage (SAH) but normal cranial computed tomography (CT). METHODS: All consecutive patients undergoing both cranial CT and lumbar puncture for investigation of possible SAH at the Princess Alexandra Hospital, Brisbane, between January 2000 and April 2003 were included in the study. All case histories, radiology and laboratory results were retrospectively assessed. The sensitivity, specificity and positive predictive value of the spectrophotometry test were calculated. RESULTS: 253 patients were included in the study. Spectrophotometry was shown to have a sensitivity of 100% but a specificity of only 75.2%. The positive predictive value of spectrophotometry as an indicator of SAH was 3.3%. CONCLUSIONS: CSF spectrophotometry has an unacceptably low specificity and positive predictive value, which greatly limit its use as a clinical tool.
AIMS: To assess the sensitivity and specificity of CSF spectrophotometry for the detection of xanthochromia in patients with clinical symptoms suggestive of subarachnoid haemorrhage (SAH) but normal cranial computed tomography (CT). METHODS: All consecutive patients undergoing both cranial CT and lumbar puncture for investigation of possible SAH at the Princess Alexandra Hospital, Brisbane, between January 2000 and April 2003 were included in the study. All case histories, radiology and laboratory results were retrospectively assessed. The sensitivity, specificity and positive predictive value of the spectrophotometry test were calculated. RESULTS: 253 patients were included in the study. Spectrophotometry was shown to have a sensitivity of 100% but a specificity of only 75.2%. The positive predictive value of spectrophotometry as an indicator of SAH was 3.3%. CONCLUSIONS:CSF spectrophotometry has an unacceptably low specificity and positive predictive value, which greatly limit its use as a clinical tool.