Predicting decannulation success in patients with neurological conditions: development and validation of the NEURODECANN Clinical Score

Decannulation is crucial in improving outcomes and quality of life for patients with tracheostomy and neurological injuries. This study aimed to develop and retrospectively assess the NEURODECANN Clinical Score to predict decannulation success. A retrospective cohort of patients with a tracheostomy (n = 117) was analyzed, distinguishing those who achieved decannulation from those who did not. Logistic regression and Random Forest analysis identified key predictors: level of consciousness, diagnosis, blue dye test, amount of secretions, speaking valve use, secretion management, dysphagia, maximum inspiratory pressure, and cough. Model performance was assessed using the area under the curve (AUC), sensitivity, and specificity. NEURODECANN demonstrated excellent predictive accuracy (AUC = 0.95, 95% CI: 0.91–0.98), with a sensitivity of 92.6% and specificity of 87.3%. Patients scoring ≥ 13 points had a high probability of successful decannulation. The NEURODECANN Clinical Score provides a structured, evidence-based framework to support clinician-driven decannulation decisions, optimizing tracheostomy management in people with neurological conditions. Further validation is needed to confirm its applicability across diverse clinical settings.