Predictors of Postoperative Cerebrospinal Fluid Leak and Infection After Endoscopic Endonasal Approach for Sellar and Suprasellar Masses

Introduction: The endoscopic endonasal approach (EEA) is a common surgical treatment for sellar and suprasellar masses. Two common complications of this procedure include postoperative cerebrospinal fluid (CSF) leak and postoperative infection. The rate of CSF leak and postoperative infection vary widely in the literature. This study evaluates the rates of postoperative CSF leak and infection after transsphenoidal surgery to identify any associated risk factors.

Methods: This is a retrospective review of consecutive patients undergoing EEA resection of a sellar or suprasellar mass at a single institution from July 2017 to March 2020. Common surgical practice included the involvement of an Otolaryngologist and Neurosurgery attending for all cases. Pre-operative antibiotics were administered within 30 minutes of incision. Patients were treated with ceftriaxone postoperatively unless contraindicated due to existing drug allergy. All closures consisted of either a nasoseptal flap or middle turbinate free mucosal autograft placement. Commonly used adjuncts to closure included fascia lata allograft, abdominal fat graft, DuraMatrix, and DuraSeal. Skull defect size was measured using a postoperative MRI/CT scan. Extent of Resection was determined based on an attending radiologist's interpretation of the postoperative contrasted MRI. Statistical analysis involved Chi-Square testing.

Results: A total of 175 cases were included in this study. Pathology consisted of the following: nonfunctioning adenomas (59.4%), secreting-adenomas (25.1%), and Rathke's cleft cysts (10.9%). One hundred thirty-four patients underwent gross total resection (76.5%). Incidence of intraoperative and postoperative CSF leak was found to be 33.7 and 2.2%, respectively. Three patients had postoperative infections (1.7%). 50% of patients with intraoperative CSF leaks underwent duraplasty. Twenty-four patients underwent abdominal fat graft placement (14%). Four patients required lumbar drains.

Intraoperative CSF leak was significantly associated with postoperative leak (p = 0.01). Sellar defect size was not associated with postoperative CSF leak (p = 0.063). There was no significant association between lumbar drain placement and postoperative CSF leak (p = 0.993). There was no significant relationship between extent of resection and postoperative CSF leak. Antibiotics were administered between 1 to 3 days postoperatively, the most common being ceftriaxone (68.3%). There was no significant relationship between antibiotic choice or duration and postoperative infection.

Conclusion: The rate of cerebrospinal fluid leak and postoperative infection was low in this single institution retrospective study in comparison to previously documented reports. Intraoperative CSF leak was significantly associated with postoperative CSF leak. In the setting of intraoperative leaks, mucosal or nasoseptal flap coupled with duraplasty, abdominal fat graft or fascia lata graft was sufficient in preventing postoperative leaks in the majority of cases. The evidence presented suggests that this low rate of CSF leaks could be correlated to the inclusion of either a nasoseptal flap or mucosal autograft in the closure of all patients. The low incidence of postoperative infection makes it difficult to assess the effect of antibiotic choice on outcome. Proceeding forward, further investigation comparing the outcomes between free mucosal graft versus nasoseptal flap placement as well as the ophthalmologic and endocrinology outcomes of these patients could provide further insight into the potential risk factors for adverse events.

Publikationsverlauf

Artikel online veröffentlicht:
15. Februar 2022