After an episode of acute respiratory failure requiring mechanical ventilation, weaning the patient from the ventilator may be difficult. Discontinuation of ventilation is estimated to take up to 40% of the total duration of ventilatory support. Approximately two- thirds of patients can be disconnected from the ventilator after a spontaneous breathing trial. Prolongation of mechanical ventilation may increase the risk of adverse events such as infections and can increase morbidity and mortality. Identifying weaning readiness early and reliably is therefore crucial. Weaning protocols developed to assist in identifying weaning readiness have been shown to shorten duration of mechanical ventilation, most notably the weaning period. Closed loop knowledge-based systems serve as a continuously applied weaning process that automatically reduce ventilatory assistance according to patient ability and indicate when the patient is ready for disconnection. No data on the use of such a computer driven system (CDS) in elderly patients requiring prolonged ventilation in weaning centers (non ICU) have been reported. The Objective of the present study is to evaluate the ability of a computer driven system to predict weaning readiness and to compare this to a physician-directed protocol. The CDS continuously adapts pressure support, gradually decreasing ventilator assistance according to patient ability, and thus indicates weaning readiness.

Patients who maintain spontaneous breathing for 7 days following weaning will be considered to be successfully weaned from mechanical ventilation.