Radiographic Enhancement of the Vascularized Nasoseptal Flap Does Not Predict CSF Leaks

Background and Purpose: Cerebrospinal fluid (CSF) leaks continue to be the most common postoperative complication in expanded endonasal skull base procedures. Currently, a multilayer closure with a vascularized nasoseptal flap is used as a standard for closure of the skull base in an effort to avoid CSF leaks. Because vascularized mucosa enhances avidly after intravenous (IV) contrast administration, we hypothesized that watertight closure of the ventral skull base would correlate with immediate postoperative radiographic enhancement of the nasoseptal flap with magnetic resonance imaging (MRI).

Material and Methods: We retrospectively reviewed a cohort of 24 consecutive patients who underwent advanced complexity, endoscopic endonasal resections of tumors and lesions. We excluded patients who underwent pituitary tumor resections. We calculated the incidence of CSF leak and measured the presence of nasoseptal flap enhancement. Additional variables of interest included age, gender, complexity of surgery, chronology of surgery, use of free fat graft, use of free fascia lata graft, use of collagen substitute, and use of a lumbar drain. This study was thus designed as a retrospective cohort study.

Results: Of the 24 patients, 19 had immediate postoperative MRIs; 3 had postoperative CSF leaks. Fifteen of 19 patients had radiographically enhancing nasoseptal flaps. Of these 15 patients, 3 developed a CSF leak. In contrast, of the four patients with no evidence of flap enhancement, none developed a postoperative CSF leak (Fisher's exact test, P = 1.0). CSF leak was associated with posterior fossa lesions (P = 0.25). Nasoseptal flap enhancement was associated with younger age (P = 0.15).

Conclusion: The vascularized pedicled nasoseptal flap has become a staple of complex endoscopic skull base closures. This retrospective cohort study fails to confirm the correlation between CSF leak and immediate, postoperative radiographic enhancement of the nasoseptal flap. Hence, this negative study calls to question the assumption that it is the vascularity of the nasoseptal flap that has resulted in lowering of overall CSF leaks in expanded endonasal skull base procedures.