Investigation of Skull-Based Cerebrospinal Fluid Leak Repair: A Single-Institution Comprehensive Study of 116 Cases Over 10 Years.

Background: Cerebrospinal fluid (CSF) leaks have been historically difficult to diagnose and treat because their cause can widely vary. There are insufficient diagnostic predictors and no clinically accepted standards for their treatment. This large institutional study reports on the diagnosis, management, and outcomes of patients presenting with CSF leak over 10 years and aims to identify potential comorbidities and risk factors for primary and recurrent leaks.

Methods: Patients diagnosed with CSF leak from 2007 to 2017 were analyzed retrospectively. The data included medical history, body mass index, surgical treatment, and postoperative outcomes.

Results: A total of 116 cases were identified. The location of leaks was 91 CSF rhinorrhea and 28 CSF otorrhea (3 both). The average BMI for females was greater than that of males (P = 0.01). Causes of leak were 64 noniatrogenic, 47 iatrogenic, and 9 traumatic. A total of 108 patients underwent surgical treatment. Sixty-nine were treated by endoscopic approaches, 42 involved open approaches, and 83 involved the placement of a lumbar drain. Eighteen patients had a ventriculoperitoneal shunt and 6 had a lumbar-peritoneal shunt. A total of 78 patients (72.22%) had an associated encephalocele with the CSF leak. The average length of stay was 7.73 days (0.76). The average length of follow-up was 1.58 years (0.22). The primary repair rate was 80.17% (n = 93) and the overall repair outcome was 99.14% (n = 115).

Conclusions: The overall CSF repair outcome was 99.14% over 10 years at a single institution. Despite this high percentage, CSF leaks continue to be a complex problem and require vigorous multidisciplinary work with close follow-up and use of multiple imaging strategies.