Retrospective validation of a β-trace protein interpretation algorithm for the diagnosis of cerebrospinal fluid leakage

Abstract

Background: Cerebrospinal fluid (CSF) leakage is a rare condition that can potentially lead to the development of serious complications. In the last decade, β-trace protein (β-TP) has been shown to be a valuable immunological biomarker that allows prompt and non-invasive identification of CSF leakage. At our institution, the measurement of β-TP has been included in the diagnostic work-up of CSF leakage for more than 10 years. According to our diagnostic algorithm, the presence of CSF in secretion is excluded when β-TP values are <0.7 mg/L, whereas β-TP values ≥1.3 mg/L indicate the presence of CSF in secretion. β-TP values between 0.7 and 1.29 mg/L indicate the presence of CSF if the β-TP ratio (β-TP secretion/β-TP serum) is ≥2. This study aimed to validate this diagnostic algorithm using clinically defined nasal/ear secretions. Methods: We performed a retrospective statistical analysis of three β-TP interpretation strategies using data of 236 samples originating from 121 patients with suspect CSF leakage received at our laboratory between 2004 and 2012. Results: The highest odds ratio was obtained when the proposed algorithm has been used for the interpretation of β-TP results, showing a sensitivity of 98.3% and a specificity of 96%. Positive and negative predictive values were 89.2% and 99.4%, respectively. Conclusions: Our data suggest that the proposed β-TP interpretation algorithm is a valuable tool for the diagnosis of CSF leakage in the clinical practice.