2026 Poster Presentations
P247: REVISION TSA NEEDS FREQUENT FLAP REQUIREMENT AND INCREASED RISK OF FLAP NECROSIS
Donghyeok Kim, MD; Gwanghui Ryu, MD, PhD; Doo-Sik Kong, MD, PhD; Won-Jae Lee, MD; Hyo-Yeol Kim, MD, PhD; Yong Gi Jung, MD, PhD; Sang Duk Hong, MD, PhD; Samsung Medical Center
Objective: Revision surgery for pituitary tumors via the transsphenoidal approach becomes technically challenging due to scar tissue and anatomical distortion from previous operations, with additional difficulties in harvesting and achieving successful vascularization of nasoseptal flap (NSF) used for reconstruction. Particularly in revision surgery for pituitary adenomas, the high incidence of intraoperative cerebrospinal fluid (CSF) leaks makes skull base reconstruction critical. This study aimed to retrospectively analyze the frequency of intraoperative and postoperative CSF leaks, NSF utilization, and flap survival rates in patients undergoing revision transsphenoidal surgery at a single tertiary medical center, to establish evidence for effective skull base reconstruction strategies in revision surgery.
Methods: Among 1,243 endoscopic skull base surgeries performed at Samsung Medical Center from April 2021 to March 2025, 85 revision cases were retrospectively analyzed. Of these, 47 cases involved revision surgery for pituitary adenomas. Surgical records and magnetic resonance imaging (MRI) were reviewed to determine the occurrence and severity of intraoperative CSF leaks and postoperative CSF leaks. Cases utilizing NSF for reconstruction were identified, and flap viability was assessed through contrast enhancement on T1-weighted MRI obtained within 48 hours postoperatively. Cases where previously harvested NSF were detached and reused during revision surgery were analyzed separately.
Results: Intraoperative CSF leaks occurred in 34 of 47 pituitary adenoma revision cases (70.8%), with 18 cases involving high-flow Grade 3 leaks. Following appropriate skull base reconstruction, postoperative CSF leaks occurred in only 2 of 47 cases (4.2%), both of which had experienced intraoperative Grade 3 leaks. NSF reconstruction was performed in 29 cases (61.7%). Contrast-enhanced MRI evaluation of flap vascularity was available in 26 cases, of which 4 (15.4%) showed no contrast enhancement, suggesting flap failure or necrosis. In 5 cases where NSF had been previously harvested during initial surgery, the same flaps were detached and reused during revision surgery. Among these, 4 cases demonstrated good contrast enhancement of the reused flaps, indicating adequate survival. None of these 5 cases developed postoperative CSF leaks.
Conclusion: In revision transsphenoidal surgery for pituitary adenomas, intraoperative CSF leaks occur more frequently than in primary surgery, particularly high-grade massive leaks. However, through meticulous skull base reconstruction using NSF, the actual incidence of postoperative CSF leaks can be maintained at low levels. The approximately 15% rate of absent contrast enhancement in NSF among revision patients suggests that flap vascularization is more frequently compromised compared to primary surgery, with increased risk of flap necrosis. This reduced NSF survival is likely due to difficulty maintaining the vascular distribution caused by scarring and bilateral adhesions of the nasal septum from previous surgery. Therefore, particular attention must be paid to preserving the sphenopalatine artery when elevating NSF during revision surgery in patients who had sphenoid rostrum removal in previous operations. Importantly, previously harvested NSF can be successfully reused with appropriate dissection and handling, and most reused flaps maintain adequate blood flow for successful skull base reconstruction.