Safety and feasibility of percutaneous tracheostomy performed by neurointensivists.

Introduction: We evaluated the effects of a change from routine surgical tracheostomy (ST), performed primarily by ENT surgeons, to bedside percutaneous tracheostomy (PT) performed by neurointensivists.

Methods: The first 67 PT procedures performed by neurointensivists were retrospectively reviewed, and compared with 68 consecutive ST procedures performed during the previous year. Demographics, severity of illness, procedural complications, incidence of ventilator-associated pneumonia (VAP), duration of mechanical ventilation (DMV), length of stay (LOS), and hospital charges were evaluated.

Results: Age, race, gender, neurological diagnoses, comorbid illnesses, and Glasgow coma scale on admission and the day of tracheostomy were similar. Procedural complications occurred in 8% of PT patients and 9% of ST patients, including clinically significant bleeding, transient loss of the airway, ICP rise requiring treatment, or acute lung injury (P = 0.3). PT was performed earlier than ST (median [interquartile range] ventilator day 8 [4-11] vs. 12 [8-18], P = 0.001). Median DMV was shorter in the PT cohort (19 [10-27] vs. 24 [16-33] days, P = 0.02), as was median ICU LOS (15 [9-21] vs. 19 [12-27] days, P = 0.01). ICU charges (US dollars) were lower in the PT cohort (median $123,404 vs. $156,311, P = 0.01). Trends toward less VAP, shorter hospital LOS, and lower total hospital charges among patients receiving PT did not achieve significance.

Conclusions: PT performed by neurointensivists was safe compared to ST. Timely PT by neurointensivists may offer significant advantages in terms of ventilator weaning, ICU LOS, and the cost of care.