The role of early tracheostomy in critically ill neurosurgical patients.

OBJECTIVE To determine the value of early tracheostomy (within 7 days) in ventilated neurosurgical patients. METHODS Retrospective review of intubated patients in the neurosurgical intensive care unit (NICU) who underwent elective open tracheostomies for prolonged ventilation. RESULTS Thirty patients over a 2-year period were analysed. There were 19 males and 11 females, mean age 53.9 +/- 18.1 years (range 14 to 89), and mean Glasgow Coma Scale (GCS) score on admission 7.1 +/- 3.8 (range 3 to 15). The underlying disease aetiology was cerebrovascular disease in 53% of patients, head trauma in 33% and tumour or infection in 13%. Tracheostomy was performed after a mean period of 8.5 +/- 3.5 days (range 2 to 18), with patients requiring ventilation for a mean duration of 13.5 +/- 6.3 days (range 3 to 31). Complications were minimal; 1 wound infection (3.3%) and 4 tube obstructions (13.3%). Patients who underwent elective early tracheostomy (Group 1 = within 7 days) had poorer GCS on admission (6.3 +/- 2.9 versus 7.7 +/- 4.3 in Group 2, P = 0.271). Tracheostomy was performed after a mean of 5.3 +/- 1.7 days in Group I vs. 10.6 +/- 2.7 days in Group 2. Group 1 patients had faster recovery from nosocomial pneumonia (12.3 +/- 6.2 versus 17.9 +/- 12.5 days, P = 0.168), shorter duration of ventilation (9.8 +/- 5.9 versus 16.0 +/- 5.4 days, P = 0.007), and reduced incidence of multibacterial tracheobronchial colonisation (42% versus 72%, P = 0.098). The most prevalent organisms were Acinetobacter baumanii (43.3%), Pseudomonas (40%), methicillin-resistant Staphylococcus aureus (MRSA) (33%), Klebsiella (30%) and Staphylococcus aureus (26.7%). CONCLUSION Early tracheostomy in selected neurosurgical patients with poor GCS scores was associated with reduced incidence of tracheobronchial colonisation by multiple pathogens, improvement in chest infections, and rapid weaning from ventilatory support.