Comparative analysis of outcomes following craniotomy and expanded endoscopic endonasal transsphenoidal resection of craniopharyngioma and related tumors: a single-institution study.

OBJECTIVE: Craniopharyngiomas and similar midline suprasellar tumors have traditionally been resected via transcranial approaches. More recently, expanded endoscopic endonasal transsphenoidal approaches have gained interest. Surgeons have advocated for both approaches, and at present there is no consensus whether one approach is superior to the other. The authors therefore compared surgical outcomes between craniotomy and endoscopic endonasal transsphenoidal surgery (EETS) for suprasellar tumors treated at their institution.
METHODS: A retrospective review of patients undergoing resection of suprasellar lesions at Cedars-Sinai Medical Center between 2000 and 2013 was performed. Patients harboring suspected craniopharyngioma were selected for extensive review. Other pathologies or predominantly intrasellar masses were excluded. Cases were separated into 2 groups, based on the surgical approach taken. One group underwent EETS and the other cohort underwent craniotomy. Patient demographic data, presenting symptoms, and previous therapies were tabulated. Preoperative and postoperative tumor volume was calculated for each case based on MRI. Student t-test and the chi-square test were used to evaluate differences in patient demographics, tumor characteristics, and outcomes between the 2 cohorts. To assess for selection bias, 3 neurosurgeons who did not perform the surgeries reviewed the preoperative imaging studies and clinical data for each patient in blinded fashion and indicated his/her preferred approach. These data were subject to concordance analysis using Cohen's kappa test to determine if factors other than surgeon preference influenced the choice of surgical approach.
RESULTS: Complete data were available for 53 surgeries; 19 cases were treated via EETS, and 34 were treated via craniotomy. Patient demographic data, preoperative symptoms, and tumor characteristics were similar between the 2 cohorts, except that fewer operations for recurrent tumor were observed in the craniotomy cohort compared with EETS (17.6% vs 42.1%, p = 0.05). The extent of resection was similar between the 2 groups (85.6% EETS vs 90.7% craniotomy, p = 0.77). An increased rate of cranial nerve injury was noted in the craniotomy group (0% EETS vs 23.5% craniotomy, p = 0.04). Postoperative CSF leak rate was higher in the EETS group (26.3% EETS vs 0% craniotomy, p = 0.004). The progression-free survival curves (log-rank p = 0.99) and recurrence rates (21.1% EETS vs 23.5% craniotomy, p = 1.00) were similar between the 2 groups. Concordance analysis of cases reviewed by 3 neurosurgeons indicated that individual surgeon preference was the only factor that determined surgical approach (kappa coefficient -0.039, p = 0.762)
CONCLUSIONS: Surgical outcomes were similar for tumors resected via craniotomy or EETS, except that more CSF leaks occurred in the EETS cohort, whereas more neurological injuries occurred in the craniotomy cohort. Surgical approach appears to mostly reflect surgeon preference rather than specific tumor characteristics. These data support the view that EETS is a viable alternative to craniotomy, providing a similar extent of resection with less neurological injury.