Increasing focus on the negative effects of bed rest have become more apparent in the intensive care unit within the last decade. A few studies have found an association between early rehabilitation starting at the intensive care unit and outcome after discharge from rehabilitation. The early mobilization presents with challenges regarding haemodynamic stability. The aim of this trial is to assess the feasibility before conducting a larger randomised trial that will investigate benefits and harms of an intensive physical rehabilitation intervention focusing on mobilisation to the upright position, starting as early as clinically feasible in the intensive care unit

Patients with severe acquired brain injury (ABI) may benefit from early and intensive rehabilitation that in part consists of physical exercise. Studies have found an association between higher-level physical activities and final outcome in patients with ABI. A higher-level activity for patients with severe ABI and disorders of consciousness consists primarily of mobilization to upright standing position on a tilt-table. Intense higher level activities of severe ABI patients are usually not initiated in the acute stage after injury but rather at a later, sub acute stage (weeks), when the patients have been stabilized and transferred to a highly specialized rehabilitation unit. Andelic et al (2012) conducted a non-randomized cohort study where they investigated the effects of early rehabilitation at the intensive care unit. Although the consistency of the rehabilitation paradigm was unspecified, they did observe a benefit of early intervention. A recent randomized pilot study in 31 patients with acute severe ABI and disorders of consciousness concluded that early mobilization using a tilt-table with integrated stepping that increases the venous return of blood to the heart, could be conducted safely, with significant improvements after three months. The hypothesis that early mobilization of patients with severe traumatic brain injury leads to better functional outcome at discharge from the rehabilitation unit and at one year post injury compared to patients that receive usual care needs to be tested in a larger clinical trial.

In parallel, the physiological changes the patients experience due to their injury and the wast amount of bed rest and the possible association with the patients' clinical outcome are explored. Training is often limited by orthostatic intolerance. The physiological mechanisms causing orthostatic hypotension and their recovery have not been thoroughly investigated. In other patient populations with neurally mediated syncope or orthostatic hypotension, intensive tilt-table training has been shown to be beneficial. In addition, recent studies including a large number of ABI patients have found an association between impaired cerebral autoregulation measured the first days after injury and an unfavorable outcome.

Therefore, we wish to assess the feasibility of an early head-up tilt protocol in patients with severe TBI, not only in terms of the number of patients that are successfully mobilised, but also of the number of adverse events and reactions. In exploratory analyses, we will assess physiological outcomes within the first four weeks and clinical outcomes at three months and one year.

The intervention group receives an early and intensive mobilization programme with head-up tilt, during their stay in the intensive care unit and throughout the early stages of rehabilitation. Mobilization will be conducted using a tilt-table with integrated stepping (The ERIGO® from HOCOMA company in Switzerland). The programme will be conducted as a supplement to the patient's usual care.The tilt-table intervention is applied five times per week for a maximum of four weeks during the stay in the neurointensive care unit. Each session consists of 20 min. mobilization. Within each session the patient will be moved to the tilt-table and secured with straps and harness. The patient is then mobilized step wise to 30°, 50° and 70° head-up tilt in one min. intervals while blood pressure, heart rate, and respiratory rate are closely monitored. Cerebral perfusion pressure and intracranial pressure are monitored if relevant. If at any time the predetermined safety limits for blood pressure, cerebral perfusion pressure, intracranial pressure or heart rate are violated, the patient is lowered to 0° tilt (supine position). This procedure is continued until the patient has been tilted upright for a maximum of 20 min. or until a total duration of 40 min. for the head-up tilt procedure has been reached.

If the patient is discharged from the intensive care unit before four weeks, training will continue at the department of neurorehabilitation with a pre-specified tilt-table protocol consisting of mobilization twice a day on a similar tilt-table. Occasionally patients will be transferred to an intensive care unit at another hospital, while waiting for a further training at the department of neurorehabilitation. These patient will continue their mobilization programme on a regular tilt-table without active stepping. Patients who show functional improvement beyond the scope of tilt-table training (e.g. are able to stand from a chair) before the study period has ended, will have their final evaluation performed immediately hereafter and subsequently the standard rehabilitation regimen will be continued.

The control group receives standard care consisting of interdisciplinary rehabilitation. A very small part of the standard care consists of mobilizing the patient to the edge of the bed or to a wheelchair.

At inclusion the patients will be randomized to either group through an open ended blinded randomization procedure, with stratification according to the patients Glasgow Coma Score at the time (3-6 or 7-10). The randomization will consist of blocks of random sizes.

Assuming that the normality assumption is not violated the functional scores and the physiological data will be analysed with analysis of variance (ANOVA) or other linear regression models that takes into account more than two measures over time. Between-group analysis of demographic data will be performed using Student's t test with unequal variance for analysis of two groups or the chi-square test for nominal data.

Inclusion Criteria:

• Traumatic brain injury (TBI)
• Disorders of consciousness (with a tentative diagnosis of the vegetative or minimally conscious state), with a Glasgow Coma Score < 10 during wake-up call.
• Stable intracranial pressure (ICP < 20 mmHg for 24 hours).
• Must be able to mobilise beyond 30 degrees elevation

Exclusion Criteria:

• Unstable fractures contraindicating mobilisation.
• Known heart disease or liver cirrhosis prior to brain injury.
• Spinal cord injury.