RESUMO
BACKGROUND: Determining prognostic factors for the probability of tracheostomy decannulation is key to an adequate therapeutic plan. METHODS: A retrospective cohort study of 160 paediatric patients undergoing tracheostomy was conducted. Associations between different parameters and eventual tracheostomy decannulation were assessed. RESULTS: Mean follow-up duration was 27.8 months (interquartile range = 25.5-30.2 months). Median age at tracheostomy was 6.96 months (interquartile range = 3.37-29.42 months), with median tracheostomy maintenance of 14.5 months (interquartile range = 3.7-21.5 months). The overall tracheostomy decannulation rate was 22.5 per cent. Factors associated with a higher probability of tracheostomy decannulation included age at tracheostomy (hazard ratio = 1.11, 95 per cent confidence interval = 1.03-1.18) and post-intubation laryngitis as an indication for tracheostomy (hazard ratio = 2.25, 95 per cent confidence interval = 1.09-4.62). Neurological (hazard ratio = 0.30, 95 per cent confidence interval = 0.12-0.80) and pulmonary (hazard ratio = 0.41, 95 per cent confidence interval = 0.18-0.91) co-morbidities were negatively associated with tracheostomy decannulation. The probability of tracheostomy decannulation decreased significantly with increasing numbers of co-morbidities (p < 0.001). CONCLUSION: Age, post-intubation laryngitis, and number and type of co-morbidities influence tracheostomy decannulation rate in the paediatric population.