Intraoperative and Postoperative Cerebrospinal Fluid Leaks in Endoscopic Transsphenoidal Pituitary Surgery

Objective: Cerebrospinal fluid leakage, both intraoperative and postoperative, is a common complication of endoscopic transsphenoidal pituitary surgery. Intraoperative leaks are significantly more frequent, with reported rates ranging from 16 to 30% as compared with postoperatively ranging from 0.6 to 16.9%. Herein, we attempt to identify risk factors for intraoperative and postoperative leaks after endoscopic transsphenoidal pituitary surgery.

Methods: A total of 346 patients with pituitary adenomas treated by endoscopic transsphenoidal surgery at a single tertiary referral institution were identified between 2013 and 2017. A retrospective chart review was performed.

Results: Overall intraoperative and postoperative leak rates were 29.8 and 1.73%, respectively. 2.9% of patients with an intraoperative leak developed a postoperative leak. In 299 patients who underwent primary surgery, the intraoperative leak rate was 28.4% and postoperative leak rate was 1.0%. In 62 patients undergoing secondary surgery, the intraoperative leak rate was 35.5% and postoperative leak rate was 4.8%. Having undergone prior pituitary surgery was not associated with a statistically significant increased rate of intraoperative or postoperative leaks.

In our series, 6 patients had a postoperative CSF leak. The mean length of time before developing a postoperative leak was 9.8 days. All patients with postoperative leakage were treated surgically and in 50% a lumbar drain was also placed. Initial repair techniques used were the nasoseptal flap (50%), synthetic material alone (33%) and Surgicel™ repair alone (16.7%). Presence of a high flow leak intraoperatively was significantly associated with postoperative leak (p < 0.01). Of 8 patients with a high flow leak, 3 were repaired with a nasoseptal flap and developed a postoperative leak. Cavernous sinus invasion were significantly associated with intraoperative leaks (p < 0.05), but not postoperative leaks. BMI, age, gender, and tumor dimensions were not associated with intraoperative or postoperative leaks. No cases of meningitis were reported.

Conclusion: Our intraoperative and postoperative leak rates for endoscopic transsphenoidal resection of pituitary adenomas are consistent with the existing literature. High flow nature of intraoperative leaks was significantly associated with the development of a postoperative leak. All postoperative leaks occurred early on and were successfully repaired with surgery.

Die Autoren geben an, dass kein Interessenkonflikt besteht.