J Neurol Surg B Skull Base 2016; 77 - A078
DOI: 10.1055/s-0036-1579866

Cerebrospinal Fluid Leak after Microvascular Reconstruction of Large Craniofacial Defects with Orbital Exenteration

Kurren S. Gill 1, David Hsu 1, Patrick Tassone 1, Howard Krein 1, Jurij Bilyk 1, Ann Murchison 1, Sara Lally 1, Ryan Heffelfinger 1, Marc Rosen 1, James J. Evans 1, Gurston Nyquist 1, Joseph Curry 1, John Pluta 1
  • 1Thomas Jefferson University, Philadelphia, Pennsylvania, United States

Background: Extensive craniofacial resections may require orbital exenteration, which carries a risk of cerebrospinal fluid (CSF) leak.

Objective: To assess the risk factors for postoperative complications, specifically CSF leak, after resection and microvascular reconstruction for orbitofacial and craniofacial defects requiring orbital exenteration (CFOE).

Methods: A retrospective review of 70 consecutive patients that underwent microvascular reconstruction for CFOE defects between September 2006 and December 2014 was performed. Data reviewed included demographics, surgical defect characteristics, type of free flap used, length of stay (LOS) and follow-up, and complications including CSF leaks.

Results: Thirty-two patients had exposure of the orbital apex or orbital roof alone. Fifteen patients had dura exposed and 23 had dura resected. Of the dural exposure / resection group, 24 had anterior cranial fossa defects, 5 had middle cranial fossa defects, and 9 had defects of both fossae. The overall complication rate was 31.6%. Five patients developed a postoperative CSF leak, and 3 required operative management. All 5 patients with a CSF leak had middle fossa exposure and previous radiation and 4 had previous surgery. No patients with exposure / resection of anterior fossa alone or orbital apex exposure had a leak. On univariate analysis, middle fossa dural exposure / resection and previous radiation showed a statistically significant risk for CSF leaks (p = .008 and p = 0.0001, respectively). Multivariate analysis revealed location to be the only significant predictor of CSF leak (p = 0.003). There was evidence to suggest a relationship between previous radiation and CSF leak, though the result was not statistically significant (p = 0.06). Previous surgery was similarly found to be independent of CSF leak (p = 0.627). Additionally, no significant association with complications was seen with type of free flap used (p = 1) or with the blood supply used for flap anastomosis.

Conclusion: In CFOE, middle cranial fossa exposure and previous irradiation increase the risk of CSF leak after reconstruction, while exposure of the orbital apex or anterior fossa confer a lower risk for CSF leak.