[A study of early tracheostomy in patients undergoing prolonged mechanical ventilation]

Several complications are associated to mechanical ventilation (MV), especially if the duration of MV is prolonged: nosocomial pneumonias and sinusitis, laryngeal and tracheal complications (such as stenosis or granuloma). Pneumonias, en particular, seem to be associated to an increased mortality risk. Overall, the hospital mortality rate of patients undergoing MV for more than 7 days is 45%. Early tracheostomy could allow to reduce some of these complications; however, the actual value of this procedure has never been proven.

The aim of this multicenter randomized trial is to assess the potential interest of early tracheostomy (before the 4th day of MV) compared to translaryngeal intubation, in critically ill patients undergoing MV for more than 7 days, in reducing mortality rate, incidence of pneumonias, and duration of MV.

Inclusion criteria are: MV since<4 days, forseable duration of MV>7 days, age>18-yr, and informed consent obtained. Exclusion criteria are: MV since > 4 days, presence of tracheostomy, major risk of bleeding, cervical infectious disease, "moribund" state (according to SAPS II and OSF scores), MV or CPAP at home, non reversible neurologic disease, intracranial hypertension. The randomization is performed before the end of the 4th day of MV.

A decrease of mortality rate on d.28 from 45% to 32% is expected (two-sided test, alpha=0.05, B=0.80); overall, 468 patients should be included over a 3-yr period in at least 50 centers. In addition, a decrease of the incidence of pneumonias and of duration of MV during the first 28 days is expected. Secondary endpoints are mortality rate on d.60, hospital mortality, total duration of MV, infectious complications (other than pneumonias), laryngeal and tracheal complications, duration of sedation, duration of stay in intensive care unit, hospital costs, and comfort of patients.