Outcome of Endoscopic Transsphenoidal Surgery for Recurrent or Residual Pituitary Adenomas and Comparison to Non-Recurrent or Residual Cohort by Propensity Score Analysis

ObjectiveTo evaluate the long-term outcomes and safety of endoscopic transsphenoidal surgery (ETS) in recurrent and residual pituitary adenomas (rrPAs), as well as the predictors of gross total resection (GTR) and intraoperative CSF leakage. Furthermore, to compare outcomes and complications with non-rrPAs cohort. MethodsClinical and radiological characteristics of patients with rrPAs who underwent ETS were collected between 2017 and 2020. Data of patients with non-rrPAs were collected from 2019 to 2020. Logistic regression analyses were performed to investigate the factors influencing gross total resection (GTR) and intraoperative CSF leakage. Between-group comparisons of outcomes and complications were performed through propensity score analysis. ResultsWe enrolled 73 patients with rrPAs. GTR was achieved in 41 (56.1%) cases; further, GTR or near-total resection was achieved in 93.2% of patients. The mean tumor volumes for GTR and non-GTR cases were 6.2 +/- 7.2 cm(3) and 11.1 +/- 9.1 cm(3), respectively. Multivariate regression analysis of the GTR rate in patients with rrPAs revealed that Knosp grade was an independent factor (odds ratio [OR] = 0.324; p=0.005). Moreover, previous transcranial surgery and non-functional pituitary adenomas were risk factors for intraoperative CSF leakage in patients with rrPAs (OR=6.450, p=0.019 and OR=7.472, p=0.012, respectively). After propensity score matching, There was no significant difference in the GTR rate between patients with rrPAs and patients with non-rrPAs. Contrastingly, patients with rrPAs had a higher rate of intraoperative CSF leakage and longer postoperative hospital stay than patients with non-rrPAs. During the follow-up, vision improved in 9 (22.0%) and 24 (62.5%) patients with rrPAs and non-rrPAs, respectively. Although there was a trend that reoperation of rrPAs involved a lower hypopituitarism recovery rate and biochemical remission rate, as well as a higher hypopituitarism rate, there was no statistically significant between-group difference. ConclusionsKnosp grade was an independent factor for GTR in endoscopic transsphenoidal surgery in patients with rrPAs. Previous transcranial surgery and non-functional PAs were risk factors for intraoperative CSF leakage. Although associated with longer hospital stay, rrPAs did not associate with lower GTR rate or more frequent postoperative complications than non-rrPAs cohort.