Scientific Reports (Sep 2024)

Risk factor analysis and prediction model to establish recurrence or progression of non-functioning pituitary adenomas in men after transnasal sphenoidal surgery

  • Jiansheng Zhong,
  • Yuyang Chen,
  • Mingyue Wang,
  • Jun Li,
  • Ziqi Li,
  • Haixiang Li,
  • Liangfeng Wei,
  • Shousen Wang

DOI
https://doi.org/10.1038/s41598-024-72944-5
Journal volume & issue
Vol. 14, no. 1
pp. 1 – 13

Abstract

Read online

Abstract This paper aims to analyze the risk factors for the recurrence or progression of non-functioning pituitary adenomas (NFPAs) in male patients after transnasal sphenoidal surgery and to develop a predictive model for prognosis. Clinical and follow-up data of 126 male patients with NFPAs treated by transnasal sphenoidal surgery from January 2011 to January 2021 in Fuzhou 900th Hospital were retrospectively analyzed. Lasso regression analysis was used to screen the best predictors, and the predictors were further screened by multivariate logistic regression analysis, and the nomogram prediction model was constructed. The performance of the model was verified by three aspects: discrimination, calibration and clinical utility by using the consistency index (C-index), receiver operating characteristic curve (ROC), calibration curve, clinical decision curve (DCA) and Clinical impact curve (CIC). Out of 126 cases, 7 (5.56%) showed postoperative tumor recurrence, and 18 (14.29%) exhibited postoperative residual regrowth (progression). Age (P = 0.024), maximum tumor diameter (P < 0.001), modified Knosp grade (P < 0.001), resection extent (P < 0.001), Ki67 (P < 0.001), pressure symptom (P < 0.001), Pre-op hypopituitarism (P = 0.048), Post-op new hypopituitarism (P = 0.017) showed significant differences among the recurrence group, the progression group, and the alleviation group. Three independent risk factors (Ki67, modified Knosp grade, and resection extent) affecting postoperative remission were used to construct a predictive model for long-term postoperative failure to remit. The C-index of the nomogram model was 0.823, suggesting that the model had a high discriminatory power, and the AUC of the area under the ROC curve was 0.9[95% CI (0.843, 0.958)]. A nomogram prediction model based on modified Knosp grading (grades 3B-4), resection extent (partial resection), and Ki-67 (≥ 3%) predicts the recurrence or progression of NFPAs in men after transnasal sphenoidal surgery.

Keywords