AUTHOR=Chen Xiao-Yong , Chen Yue , Lin Ni , Chen Jin-Yuan , Ding Chen-Yu , Kang De-Zhi , Wang Deng-Liang , Fang Wen-Hua
TITLE=A Nomogram for Predicting the Need of Postoperative Tracheostomy in Patients With Aneurysmal Subarachnoid Hemorrhage
JOURNAL=Frontiers in Neurology
VOLUME=12
YEAR=2021
URL=https://www.frontiersin.org/journals/neurology/articles/10.3389/fneur.2021.711468
DOI=10.3389/fneur.2021.711468
ISSN=1664-2295
ABSTRACT=Objective: Early identification for the need of tracheostomy (TT) in aneurysmal subarachnoid hemorrhage (aSAH) patients remains one of the main challenges in clinical practice. Our study aimed to establish and validate a nomogram model for predicting postoperative TT in aSAH patients.
Methods: Patients with aSAH receiving active treatment (interventional embolization or clipping) in our institution between June 2012 and December 2018 were retrospectively included. The effects of patients' baseline information, aneurysm features, and surgical factors on the occurrence of postoperative TT were investigated for establishing a nomogram in the training cohort with 393 patients. External validation for the nomogram was performed in the validation cohort with 242 patients.
Results: After multivariate analysis, higher age, high neutrophil-to-lymphocyte ratio (NLR), high World Federation of Neurological Surgeons Scale (WFNS), and high Barrow Neurological Institute (BNI) grade were left in the final logistic regression model. The predictive power of the model was excellent in both training cohort and validation cohort [area under the curve (AUC): 0.924, 95% confidence interval [CI]: 0.893–0.948; AUC: 0.881, 95% CI: 0.833–0.919]. A nomogram consisting of these factors had a C-index of 0.924 (95% CI: 0.869–0.979) in the training cohort and was validated in the validation cohort (C-index: 0.881, 95% CI: 0.812–0.950). The calibration curves suggested good match between prediction and observation in both training and validation cohorts.
Conclusion: Our study established and validated a nomogram model for predicting postoperative TT in aSAH patients.