Novel Repair of High-Flow Skull Base Defects After Endoscopic Endonasal Tumor Resection Using A Fat-Fascia Lata “Sandwich” Graft

Introduction: Reconstructing high-flow cerebrospinal fluid (CSF) leaks after endoscopic tumor resection necessitates a multi-layer closure. Vascularized flaps are the gold standard for the onlay. However, inlay repair techniques are heterogeneous with variable efficacy. Free fat grafts can obliterate large dead spaces but there is a risk of neurovascular compression and migration into the third ventricle. Modifying the previously described button graft, we present a novel inlay technique using compressed rolled fat between two layers of fascia lata, termed a “sandwich” graft. Methods: Individuals who underwent endoscopic skull base tumor resection and repair of a high-flow leak with fat and fascia lata between January 2010 to August 2025 were included. A dural defect ≥1 cm2 or communication with a subarachnoid cistern or ventricle were considered high flow. Prior to 2020, fat and fascia were used as separate free layers; in 2020, the “sandwich” graft modification was incorporated. Rolled compressed fat is sutured between two pieces of fascia lata. The inner fascia is the size of the dural defect and the outer layer is 25-30% larger than the bony defect (see [ Figs. 1 ] and [ 2 ]). The primary outcome was post-operative CSF leak rates in the pre- and post-sandwich graft implementation periods. Data on demographics, patient and surgical factors, defect location, and tumor pathology were collected. Radiated patients were excluded. Chi-squared, two-tailed t-tests, and binary logistic regression were used for statistical analysis. Results: Overall, 55 patients had repairs with fat and fascia as separate layers and 75 patients had sandwich grafts. Pre-sandwich patients were older (47.5 vs. 36.8 years; p = 0.018) and had a slightly higher mean BMI (31.5 vs. 28.2; p 0.009). Gender and race were similar between groups. Meningiomas and craniopharyngiomas were the most common pathology and the majority of defects were transplanum or transtuberculum. There were no differences in pathology, location, revision surgery, or peri-operative CSF drainage (with lumbar drain, extraventricular drain, or shunts) frequencies between groups. There were 6 CSF leaks (10.9%) before 2020, and this improved to one CSF leak (1.3%) for the sandwich group (p = 0.017). Binary logistic regression controlling for age, gender, race, BMI, location, pathology, revision, and CSF drainage showed that sandwich grafts were independently associated with a reduced CSF leak rate (p = 0.044). Conclusion: A fat-fascia lata sandwich graft followed by a vascularized onlay is an efficacious and safe reconstruction method for high-flow defects. The addition of compressed fat to the button graft provides additional robust tissue to form a water-tight seal without risking neurovascular structure compression or fat migration into a large dead space. Fig. 1 (A) . The three layers comprising the graft: larger outer fascia layer, smaller inner fascia layer, and rolled compressed fat. (B) . Fat placed between the two fascial layers. (C) . A 4-0 vicryl suture runs through each layer and back such that the knot is tied on the larger fascia side. (D) . Composite graft with all layers seen. Fig. 2 T1 pre-contrast sagittal MRI depiction of compressed fat for reconstruction after tuberculum meningioma resection. Publication History Article published online: 27 February 2026 © 2026. Thieme. All rights reserved. Georg Thieme Verlag KG Oswald-Hesse-Straße 50, 70469 Stuttgart, Germany