The role of polysomnography in tracheostomy decannulation of children with bronchopulmonary dysplasia.

Background Patients with bronchopulmonary dysplasia (BPD) may require tracheostomy for long-term mechanical ventilation. Polysomnography (PSG) may predict successful decannulation in children, however it is unclear how this success compares with children without a PSG. To better evaluate this role, we compared decannulation outcomes between tracheostomy-dependent children with BPD who underwent PSG before decannulation to those who did not. Methods This is a retrospective cohort study between 1 January 2007 and 1 June 2017 of tracheostomy-dependent children with BPD who were clinically considered for decannulation. Patient demographics, PSG results, and medical comorbidities were abstracted from medical records and compared between groups. Decannulation outcomes were compared between children with BPD who underwent PSG before decannulation and those who did not. Results One hundred twenty-five patients with BPD were considered for tracheostomy decannulation. Forty-six (37%) had a pre-decannulation PSG while 79 (63%) did not. Nineteen (41%) patients did not undergo decannulation within 6 months of the PSG. One (3%) patient with pre-decannulation PSG failed decannulation. Four (5%) patients without pre-decannulation PSG failed decannulation. Nineteen patients with PSG and no decannulation had significantly higher obstructive apnea-hypopnea index (OAHI) (13.62 vs 2.68 events per hour, P = 0.004), higher end-tidal CO2 max (52.84 vs 48.03 mm Hg, P = 0.035), and were older at PSG (median age, 6.04 vs 4.04 years, P = 0.008). Conclusions While successful decannulation can be achieved without a PSG in some patients, PSG is a valuable tool to identify BPD patients undergoing clinical evaluation for decannulation who would benefit from treatment of OSA before decannulation.