Development and Validation of a Dynamic Nomogram Predictive Model for Early Tracheotomy in Patients with Supratentorial Deep-Seated Intracranial Hemorrhage

Abstract Objective:Tracheotomy (TT) need in supratentorial deep-seated intracranial hemorrhage patients is hard to predict. We devised a nomogram to assess the need for immediate TT upon arrival. Methods:Retrospective study from Jan 2018 to Jan 2023 at two hospitals. Data from 1394 SDICH patients were analyzed. A nomogram using WBC, PL, HR, and GCS was created. Results: AUC of 0.817 and cut-off value of 0.139 for the training set, and AUC of 0.766 with cut-off value 0.067 for the validation set. The nomogram showed good predictive accuracy and clinical utility. Conclusion: GCS, WBC, PL, and HR are valuable markers for predicting early TT in SDICH patients.