Institutional members access full text with Ovid®

Management and Implications of Intraoperative Cerebrospinal Fluid Leak in Transnasoseptal Transsphenoidal Microsurgery

Kaptain, George J MD*; Kanter, Adam S MD†; Hamilton, David K MD‡; Laws, Edward R MD§

doi: 10.1227/NEU.0b013e318207b3fc
Operative Technique

BACKGROUND: Nonvascularized autologous grafts used for sellar reconstruction in transseptal transsphenoidal surgery are commonly applied in the setting of intraoperative cerebrospinal fluid (CSF) leak and have been shown to be effective in preventing postoperative complications.

OBJECTIVE: To assess the clinical implications of intraoperative CSF leak, to evaluate the efficacy of repair techniques using autologous nonvascularized materials, and to analyze the nature and timing of failures. These data may serve as a basis for assessing the utility of innovations in techniques and implant technologies.

METHODS: A review was conducted of 257 consecutive patients who underwent transsphenoidal surgery that was complicated by intraoperative CSF leak from 1995 to 2001. Sellar reconstruction was performed with autologous materials except in reoperations in which septal materials were not available; lumbar drain catheters were used selectively.

RESULTS: Six of the 257 patients (2.3%) developed postoperative CSF rhinorrhea occurring an average of 6.6 days after surgery. All 6 underwent reoperation, with 5 of 6 managed with operative lumbar drainage. Bacterial meningitis developed in 3 of 257 (1.2%). Worsening in visual function occurred in 8 of 257 (3.1%), with 1 of 257 (0.3%) suffering from permanent worsening of visual function. Additional surgery was performed in 2 of these patients, resulting in successful reversal of visual loss. Ten of 257 patients (3.9%) developed a subcutaneous hematoma at the fat graft harvest site, with 1 patient requiring surgical re-exploration.

CONCLUSIONS: Watertight closure of the sella with autologous materials is effective in preventing postoperative rhinorrhea. Complications specific to the technique include graft site hematoma (4%) and rare instances of visual loss caused by optic nerve compression.

*Department of Neurosurgery, Hackensack University Medical Center, Hackensack, New Jersey; †Department of Neurological Surgery, University of Pittsburgh Medical Center, Pittsburgh, Pennsylvania; ‡Department of Neurosurgery, University of Maryland, Baltimore, Maryland; §Department of Neurosurgery, Brigham and Women's Hospital, Harvard Medical School, Boston, Massachusetts

Received, October 30, 2009.

Accepted, October 1, 2010.

Correspondence: George J. Kaptain, MD, North Jersey Brain and Spine Group, 680 Kinderkmack Rd, Ste 300, Oradell, NJ 07649. E-mail:

Copyright © by the Congress of Neurological Surgeons