Graded CSF Leak Repair in Endoscopic Endonasal Base Surgeries: Retrospective Analysis of a Single Surgeon's Experience with Simple and Extended Approaches to the Skull Base

Introduction: Cerebrospinal fluid (CSF) leaks remain an important concern in skull-base surgeries. An effective graded CSF leak repair for endoscopic endonasal (EEA) approaches to the sellar/parasellar region has previously been described, but there is limited reporting on efficacy for more extensive approaches. 1 We review outcomes using this graded repair across a range of standard and extended EEAs. Methods: We retrospectively reviewed patients who underwent EEA to the anterior skull base between November 2021-May 2025. Intraoperative CSF leaks were graded as 0 (no leak), 1 (small leak confirmed by Valsalva), 2 (moderate leak with diaphragmatic defect), 3 (large leak). 1 Repair materials used included collagen sponge, autologous fat graft, and a rigid buttress of septal bone or Medpor (Stryker, Kalamazoo, USA). Demographics, surgical approach, pathology, repair details and postoperative CSF leak occurrence at discharge and follow-up were collected. Fisher’s exact test or Mann-Whitney U-test were used to compare characteristics between patients who experienced CSF leak versus those who did not. A p-value of 0.05 was considered statistically significant. Results: Ninety-four patients underwent 99 surgeries. Average age was 60 years (range 44-68) and there was a slight female predominance (50/94, 53.1%). Mean body mass index (BMI) was 30.2±8.4 kg/m 2 . Thirty-four patients (36.2%) had previously undergone sinus or skull base surgery and 13 (13.8%) had prior local radiation. The most common indication for surgery was benign pituitary tumor (49/99, 49.5%), followed by nasopharyngeal carcinoma (10/99, 10.1%), meningioma (9/99, 9.1%), encephalocele (5/99, 5.1%), esthesioneuroblastoma (4/99, 4.0%), craniopharyngioma (3/99, 3.0%), and other (19/99, 19.2%). Surgical approaches included transsphenoidal (57/99, 57.6%), transpterygoid (13/99, 13.1%), transclival (11/99, 11.1%), transtuberculum/transplanum (9/99, 9.1%), and transcribriform (9/99, 9.1%). Thirty-seven patients (37.4%) had an intraoperative CSF leak. Nine (24.3%) were Grade 1, 10 (27.0%) were Grade 2, and 18 (48.6%) were Grade 3. All transtuberculum/transplanum and transcribriform approaches had intraoperative CSF leaks of at least grade 2. Immediate post-operative repair failure was noted in one patient (2.7%) who had a grade 1 leak intraoperatively and was immediately taken back to the operating room. Zero patients had CSF leak at hospital discharge, 2-week follow-up (N=96), or at 3-month follow-up (N=90). Nasoseptal flaps were used in nineteen cases (19/99, 19.2%). These were used mostly in grade 3 leaks (15/19, 79.0%) with 4 usages in Grade 2 leaks (21.0%). Conclusion: This established intraoperative CSF leak repair strategy is effective for extended endoscopic skull-base approaches. Effective wedging of the rigid buttress is paramount to success. Tailoring the repair based on leak grade can minimize morbidity and avoid unnecessary nasoseptal flap harvest, preserving this for future use. Publication History Article published online: 27 February 2026 © 2026. Thieme. All rights reserved. Georg Thieme Verlag KG Oswald-Hesse-Straße 50, 70469 Stuttgart, Germany