Construction and validation of a Nomogram model for predicting the outcome of facial nerve function after acoustic neuroma surgery

KUAI Guohu; LIAN Minghao; LI Yandong; ZHANG Gaocai; ZHU Guohua; Maimaitili Mijiti; Geng Dangmurenjiafu

doi:10.3969/j.issn.1674-3806.2025.03.12

Abstract：Objective To construct and validate a Nomogram model for predicting the outcome of facial nerve function(FNF)after acoustic neuroma(AN)surgery.Methods The clinical data of 124 patients with AN who were admitted to Department of Neurosurgery,the First Teaching Hospital of Xinjiang Medical University from September 2020 to December 2023 were collected.All the patients underwent tumor resection through the retrosigmoid sinus approach.Multivariate logistic regression analysis was used to screen factors affecting the outcome of FNF after AN surgery,and these factors were incorporated to construct a Nomogram model.Internal validation was conducted for the Nomogram model.Results Among the 124 patients with AN,the prognosis of FNF after surgery was good in 57 cases(good prognosis group)and poor in 67 cases(poor prognosis group).Compared with the good prognosis group,the poor prognosis group was older,and had larger tumor diameter,anterior tumor length and posterior tumor length,and higher proportion of patients with moderate to severe tumor adhesion to facial nerve and cystic tumor in nature,with statistically significant differences between the two groups(P＜0.05).The results of multivariate logistic regression analysis showed that older age[OR(95％ CI)=1.078(1.025-1.134)],larger tumor diameter[OR(95％ CI)=2.231(1.274-3.906)],cystic tumor in nature[OR(95％ CI)=7.601(2.212-26.124)],moderate tumor adhesion to facial nerve[OR(95％ CI)=11.182(3.023-41.364)]and severe tumor adhesion to facial nerve[OR(95％ CI)=86.653(13.414-559.774)]were independent risk factors for poor prognosis of FNF in the AN patients after surgery(P＜0.05).Age,tumor nature,tumor diameter,and the severity of tumor adhesion to facial nerve were used as 4 variables to construct the Nomogram model.The results of receiver operating characteristic(ROC)curve analysis showed that area under the curve(AUC)of the model was 0.920(95％ CI:0.869-0.971),indicating that the model had relatively high predictive efficiency.The calibration degree of the model was validated by Bootstrap resampling method(1 000 times),and the calibration curve was found to be close to the ideal curve,indicating that the model had relatively good predictive value.The analytic results of decision-making analysis curve(DAC)showed that the DAC was above the None line and All line at the threshold range of 5％-65％,suggesting that the model had relatively good clinical practicability.Conclusion The Nomogram model constructed in this study can effectively predict the outcome of FNF after AN surgery,and has clinical application value.