Interim analysis indicated that sample size of study patients had to be highly increased to obtain significant outcomes.

The purpose of this study was to determine whether computer-driven protocolized weaning is superior to physician-directed non-protocolized weaning in over-24-hours-ventilated surgical intensive care unit (ICU) patients. The main hypothesis was that weaning duration differs between both methods. Secondary hypotheses were that reintubation rate, duration of mechanical ventilation, intensive care unit (ICU) length of stay and workload for physicians and nurses differ between both methods.

Since long-term mechanical ventilation is affiliated with a bunch of complications, weaning affected patients from ventilation as soon as possible is a crucial point. In this context automated computerized systems have become a torchbearer. These systems are expected to expedite the weaning process, reduce the duration of mechanical ventilation and ICU length of stay of a given patient in comparison to conventional physician-directed weaning. A multicenter randomized trial by Lellouche et al supported these theories, using a computer-driven system to regulate Pressure Support Ventilation (PSV). This system is now commercially available as SmartCare/PS (Draeger Medical AG & Co. KG, Luebeck, Germany).

The purpose of the study on hand was to find out whether the results reported before could be repeated in a broad, surgical ICU patient group. Study design was chosen similar. In addition we investigated the workload for physicians and nurses in both study arms which has not been conducted before.

Inclusion Criteria:

• mechanical ventilation > 24 hours
• intubation OR tracheostomy
• informed consent
• 35 - 200 kg bodyweight (77,2 - 440,9 lb)
• Ramsay-Score < 3
• spontaneous breathing at a PEEP of < 10 cm H2O
• sufficient arterial oxygenation (paO2 > 55 mmHg or SaO2 > 90%)
• haemodynamic stability (< 5 µg/kg/min Dopamine)
• body temperature (rectal) max. 39 °C / 102.2 °F
• hemoglobin > 7 g/dl
• pH > 7,2