The utility of intrathecal fluorescein in cerebrospinal fluid leak repair

Abstract

OBJECTIVE To evaluate the utility of intrathecal fluorescein (IF) for intraoperative localization and successful repair of cerebrospinal fluid (CSF) leaks. STUDY DESIGN Case series with chart review. SETTING Tertiary-care medical center. SUBJECTS AND METHODS Subjects included those undergoing endoscopic CSF leak repair with or without the use of IF. Informed consent was obtained from all patients undergoing the administration of IF (total dose 10 mg). RESULTS A total of 103 patients underwent CSF leak repair, and in 47 cases (45.6%), IF was used. Patients who were administered IF were more likely to have spontaneous CSF leak etiology (61.7% vs 16.1%; P < 0.001). Of the 47 cases with IF use, fluorescein was visualized at the skull base in 31 cases (66.0%), 11 (23.4%) had visible CSF leak without fluorescein coloration, and five (10.6%) had neither clear nor fluorescein-colored CSF visualized. Sensitivity and specificity for fluorescein detection was 73.8 percent (95% confidence interval [CI] 57.7%-85.6%) and 100 percent (95% CI 46.3%-100%), respectively. The false-negative rate was 26.2 percent (95% CI 15.8%-43.5%). Localization of the leak site was greater when fluorescein-colored CSF was visualized (100% vs 81.3%; P = 0.035). When fluorescein-colored CSF was not visualized intraoperatively, recurrence rates were 31.3 percent versus 9.7 percent when fluorescein coloration was seen, although this finding was not statistically significant ( P = 0.10). CONCLUSION The use of IF facilitates the accurate localization of CSF leaks and may assist the surgeon in confirming a watertight closure. The lack of intraoperative fluorescein visualization should not rule out the presence of CSF leak, as evidenced by a false-negative rate of 26.2 percent.