The present study aims to compare the efficiency of respiratory therapy with cough assist and the efficiency of usual respiratory therapy in intensive care patients suffering of neurologic disorder and cough ineffectiveness. The investigators hypothesis is that cough assist is more efficient than usual respiratory care in this group of patients.

Inclusion Criteria:

• Peak flow < 270 ml if the patient has one of the following diagnostic: post neurosurgery status, cerebral trauma, tetraparesia or tetraplegia

Exclusion Criteria:

• Death expected in the following 24h
• Bulbar dysfunction if extubated
• pneumothorax
• intracerebral mass effect
• elevated intracranial pressure