Modified graded skull base reconstruction for intraoperative CSF leak repair in endoscopic endonasal surgeries: a single-surgeon experience in initial years of practice and nuances in the early learning curve.

Abstract

Objective: The purpose of this study was to evaluate the outcome of a modified graded reconstruction technique based on the size of intraoperative CSF leaks in patients undergoing endoscopic endonasal surgery performed by a single surgeon in the early years of his practice.

Methods: The database of patients who underwent endoscopic endonasal approaches (EEAs) between September 2020 and August 2024 was included. Surgical complexity was categorized into levels II, III, and IV. Intraoperative CSF leak was categorized into 4 grades (0-3). Patients were divided into 2 timeline groups (those undergoing an EEA between 2020 and 2022 [group A] and between 2023 and 2024 [group B]) to assess the trends in surgical complexity and repair outcomes.

Results: A total of 69 patients with a mean age of 56 ± 16.9 years (range 12-83 years) were identified; 34 (49.3%) were female. The median body mass index was 31 (> 25 in 82.6%). The most common pathology was nonsecretory macroadenoma (57.9%). The EEA at complexity levels II, III, and IV was performed in 36.2%, 46.4%, and 17.4% of the patients, respectively. Intraoperative CSF leaks grades 1, 2, and 3 were encountered in 39.1%, 7.2%, and 17.4% of the patients, whereas 36.2% did not develop leaks (grade 0). Fat grafts and collagen matrix were used for all patients with grades 1-3. Patients with complexity level II only developed grade 1 and 2 leaks, and none of level III developed grade 3. A nasoseptal flap was used in 4 patients (5.8%), with all having level IV surgery and grade 3 intraoperative leak. Only 1 patient (1.4%) developed a postoperative CSF leak, and a lumbar drain was only used for this patient (1.4%) at the revision surgery. The rate of grades 2 and 3 leaks in group B was significantly lower than in group A (8.3% vs 33.3%, p = 0.022). Temporary nasal packing usage was also significantly lower in group B (8.3%) than in group A (28.9%) (p = 0.049), whereas high-complexity EEA rates and pedicled flap usage were not correlated with the year of surgery.

Conclusions: A graded endoscopic endonasal repair protocol, combined with the judicious use of lumbar drains and nasoseptal flaps rather than a reflexive approach, helps in minimizing postoperative CSF leak rates. Cross-training of neurosurgery graduates focusing on skull base practice is highly recommended for maximizing good outcomes in their early years of practice. With growing experience, intraoperative leak rates tend to decrease, and the reconstruction relies on a tailored multilayer strategy rather than bulky synthetic materials.