Reconstructive Techniques and Outcomes in Endoscopic Endonasal Transcribiform Resections

Introduction: Endoscopic endonasal techniques are increasingly utilized to resect tumors of the cribiform region; these resections often result in high-flow cerebrospinal fluid leaks and can have challenging defects to reconstruct (Fig. 1A-B). Various techniques have been described for anterior cranial fossa defect reconstruction broadly; however, there is a lack of literature evaluating these specific cribiform defects in terms of their reconstructive strategies and outcomes. Methods: We identified patients at our institution undergoing endoscopic endonasal resection of tumors requiring resection of dura in the cribiform region. Demographic and clinical data was collected via chart review. Multivariate logistic regression was performed with several variables (age, sex, race, smoking status, concurrent diabetes, BMI, benign vs. malignant tumor, and primary vs. revision) to determine which patients were more likely to receive more aggressive reconstruction (utilizing thigh fat and fascia vs. dural substitute). Negative binomial regression was used to analyze if hospital length of stay was associated with this more aggressive reconstruction. Results: Fifty-one patients were identified with endoscopic transcribiform resections from 2009 to 2025, 61% (n=31) female and with median age of 56.3 years. Thirty-one (61%) of patients had esthesioneuroblastomas, 11 (22%) had olfactory meningiomas, and the remaining 9 (22%) had other sinonasal tumors. All cases had intraoperative CSF leaks. For dural reconstruction, the majority of defects (n=29, 57%) were repaired with a combination of thigh fascia lata and fat as a dural inlay and overlay (Fig. 2A); less frequently, (n=17, 33%) synthetic dural matrix was instead utilized (Fig. 2B). Rarely was abdominal fat solely used as a subdural layer (n=1, 2%) or no inlay material utilized (n=4, 8%). Nasoseptal flaps were used in 48 cases (94%) with free mucosal grafts for 2 patients (4%) and one patient with a pericranial overlay (2%). There were no postoperative CSF leaks. Patients with benign tumors were more likely to have fat and fascia reconstruction compared to patients with malignant tumors, as were male patients (p=0.03, 0.036). Patients with fat and fascia reconstruction did not have a longer length of stay than other types of dural reconstruction (p > 0.05). Discussion: Our institution had success with both dural substitutes and autologous grafts. We suspect benign tumors were more likely to have this fat and fascia reconstruction given these may have been tumors with larger intracranial components (e.g. olfactory meningioma) and thus dead space in contrast to malignant sinonasal tumors with minimal intracranial component. Conclusion: Our institution has shown success in reconstruction of cribiform defects with a combination of fat and fascia or with dural matrix underlay with overlying nasoseptal flap. Fig. 1 A-B: Defects after transcribiform resection. (A) Left cribiform defect (D) after unilateral resection. OM-Contralateral (right) olfactory mucosa, L (lamina). B: Midline cribiform defect (D) after biltateral cribiform resection. Fig. 2 A-B: Underlay options in cribiform defects. A: Fascia (FA) and fat (FT) used as an underlay in a left cribiform defect. OM-Contralateral (right) olfactory mucosa, L (lamina). B: Synthetic dural matrix (DM) utilized as an underlay in a bilateral cribiform defect. Publication History Article published online: 27 February 2026 © 2026. Thieme. All rights reserved. Georg Thieme Verlag KG Oswald-Hesse-Straße 50, 70469 Stuttgart, Germany