1,000 Consecutive Endonasal Endoscopic Skull Base Cases: How Long is the Learning Curve?

Introduction: Endoscopic endonasal approaches (EEA) to the skull base have evolved over the past 20 years to become an essential component of a comprehensive skull base practice at most major academic medical centers. Given its relative infancy, large series of cases with detailed analysis of the indications, results and complications of this rapidly changing field are lacking. Many case series show a learning curve from the earliest cases, where the authors were inexperienced or were not using advanced closure techniques. We eliminated early cases to examine the alterations in practice and outcomes in a more mature series.

Methods: We reviewed a prospectively acquired database of all EEA cases performed by the senior authors at Weill Cornell Medicine/New York Presbyterian Hospital. Although the first cases were performed in 2004, we eliminated the first 200 cases and included 1,000 consecutive cases from January 2008 to December 2018 to avoid the bias created by the early learning curve. Data entered prospectively into the database included pathology, approach, closure technique and materials, intraoperative CSF leak, use of lumbar drainage and post-operative CSF leak. Other demographics and extent of resection were determined retrospectively by chart and MRI review.

Results: Of the 1,000 cases, the most common pathologies included pituitary adenoma (51%), meningoencephalocele or CSF leak repair (8.6%), meningioma (8.4%), craniopharyngioma (7.3%), odontoid (3.1%), RCC (2.8%), chordoma (2.4%), other malignancies (2.3%), and metastasis (1.6%). More procedures for craniopharyngioma (56%), odontoid (81%), RCC (68%), and metastasis (70%) were performed in the latter half of the cohort compared with the first half (44, 19, 32, and 30%, respectively). Lumbar drains were used in 24% of our cases, 32% in the first half and 16% in the second half of cases (p < 0.05). GTR increased from 40% in the first half to 73% in the second half (p < 0.005). GTR increased most dramatically for chordoma (56 vs. 100%; p < 0.05) and craniopharyngioma (41 vs. 71%; p < 0.05). The rate of any complication was 6.4% in the first half and 6.2% in the latter half of cases and vascular injury occurred in only 0.3% of cases. Post-operative CSF leak occurred in 2% of cases (meningioma 7.1%, chordoma 4.2%, RCC 3.6%, craniopharyngioma 2.7%, pituitary adenoma 0.8%) and was overall unchanged between the first and second half of the series.

Conclusion: EEA for a variety of skull base pathologies is becoming increasingly safe and effective over time. With practice, experience and an algorithm for closure based on pathology, location and presence of a leak, rates of CSF leak and complications can be quite low. Nevertheless, even after several hundred cases, there are noticeable improvements in outcome, particularly extent of resection for more complex tumors indicating a long, albeit flatter learning curve.