Abstract
Objective
Endoscopic endonasal surgery (EES) for recurrent or residual pituitary neuroendocrine tumors (rrPitNETs), particularly those involving multiple anatomical corridors, remains clinically challenging. This study evaluates the efficacy of endoscopic endonasal surgery for rrPitNETs in terms of gross total resection (GTR) rates and complications, and identifies independent predictors of resection outcomes, to provide a reference for patient selection, intraoperative decision‑making, and perioperative management.Methods
We retrospectively analyzed 152 patients with rrPitNETs undergoing EES at our institution from June 2016 to June 2024 with a median follow-up of 36 months (range, 12-96). Data included prior surgical history, clinical presentations, intraoperative techniques, perioperative complications, and long-term outcomes. Univariate and multivariate logistic regression models were applied to determine factors influencing GTR.Results
The patient cohort comprised 101 patients with one prior surgery (66.4%), 32 with two (21.1%), and 19 with three or more (12.5%). GTR was achieved in 105 patients (69.1%). Preoperative visual field defects improved in 74 of 102 patients (72.5%). Postoperative complications included new hypopituitarism in 12 patients (7.9%), permanent diabetes insipidus (DI) in 2 patients (1.3%), transient DI in 15 patients (9.9%), all resolved before discharge, delayed cerebrospinal fluid (CSF) leak in 2 patients (1.3%), managed by lumbar drainage, transient cranial nerve palsy in 13 patients (8.6%), and intracranial infection in 5 patients (3.3%), controlled with antibiotics. Multivariate analysis (separate models for tumor diameter and volume) showed that Knosp grade 3-4 had a very strong association with incomplete resection (both models, OR > 86, p Conclusions
In this selected cohort of rrPitNET patients suitable for repeat EES after structured preoperative evaluation, we observed a GTR rate of 69.1% and a major complication rate of a low major complication rate. Knosp grade 3-4 showed a very strong association with incomplete resection, although the magnitude of this effect is imprecise due to sparse data. This finding suggests that GTR is more difficult to achieve in such cases, underscoring the need for advanced preoperative planning.