Background: The recommended spectrophotometric scanning for diagnosis of subarachnoid haemorrhage (SAH) is relatively expensive, not fully automated and often requires expert interpretation. Analysis of cerebrospinal fluid (CSF) bilirubin offers an alternative approach and may obviate the need for scanning and, hence, we undertook a prospective validation study.
Methods: CSF bilirubin and spectrophotometric scans were compared from 193 patients in our institution over a two-year period and for this 162 clinical records (83.9%) were reviewed. CSF bilirubin was tested for its ability to predict elevated net bilirubin absorbance by spectrophotometry. Final outcomes were obtained by review of clinical records and contacting the New Zealand Health Information Service.
Results: From receiver operating characteristic curve analysis, there was an area under the curve of 0.99 (95% confidence interval: 0.97-1.00). At a cut-off of 359 nmol/L, corresponding to our upper reference interval, CSF bilirubin assay had a sensitivity of 100%, a specificity of 92.2% and a negative predictive value of 100%. There were no outcomes to suggest that SAH had been missed in any case.
Conclusion: CSF bilirubin is a robust screening test that accurately identifies those samples needing to be scanned and eliminates the need for this to be done in most others.