Restrictive Use of Restraints and Delirium Duration in ICU (R2D2-ICU)

The R2D2-ICU study will be a prospective, parallel-group, open label, multicenter (6 centers) randomized controlled trial. All consecutive eligible patients will be included. Patients will be randomly assigned (1/1 ratio) to either systematic PR use (systematic use group) or restrictive PR use (restrictive use group). Patients in the restrictive PR group will be subjected to PR only in case of severe agitation defined by a RASS ≥ +3. Physical restraint will consist of wrist straps. In both groups, patients will receive standardized management for analgesia, sedation, delirium detection, weaning and early mobilization according to current guidelines. Concealment will be obtained using a computer-generated randomization scheme of various-sized blocks stratified by center, age (< or ≥ 65 years) and coma at the beginning of invasive mechanical ventilation (D0)) through a centralized 24h/24h internet service. Investigation blinded to group assignment is not feasible. In both arms, patients' arousal will be evaluated twice a day until day 14 with the use of RASS. Patients with a RASS of -5 or -4 will be considered comatose (and will not be assessed for delirium). Patients with a RASS score ≥ -3 will be assessed for delirium with the use of the CAM-ICU scale twice a day.

Physical restraint (PR) is defined by the HAS (Haute Autorité de Santé française) as the use of any intervention that prevent or limit movements capability, to protect a person from an inappropriate or a dangerous behavior. In French intensive care units (ICUs), wrist straps (and sometimes ankle straps) are frequently used to prevent self-removal of medical devices in case of agitation, which can affect about 50% of patients. A survey conducted in 121 French ICUs showed that in 82 % of ICUs, PR is used at least once during mechanical ventilation (MV) in more than 50 % of patients. In 65 % of ICUs, when PR is used, it is applied for more than 50 % of MV duration. In 29 % of ICUs, PR is used in more than 50 % of cases in awake, calm and co-operative patients. PR is started without written medical prescription in more than 50 % of patients in 68 % of ICUs. Only 21 % of ICUs have a written local procedure for PR use. In the European PRICE study (566 patients, 34 ICUs, 9 countries), the frequency of PR use varied from 0% to 100%. The benefit of PR is not clearly established and PR could also be deleterious in this context. First, PR may leave patients with moderate to extremely stressful memory. Second, there is no demonstrated relationship between the PR rates and self-removal of medical devices. Finally, there is a complex relationship between agitation, PR use and delirium. While PR is prescribed to avoid potential risks associated with agitation, it seems to favor delirium. Delirium is a serious event in ICU patients that is independently associated with adverse outcomes in patients receiving MV. Moreover, the number of days of ICU delirium is significantly associated with increased 1-year mortality and long-term cognitive impairment in survivors. The R2D2-ICU study will be a prospective, parallel-group, open label, multicenter (6 centers) randomized controlled trial. All consecutive eligible patients will be included. Patients will be randomly assigned (1/1 ratio) to either systematic PR use (systematic use group) or restrictive PR use (restrictive use group). Patients in the restrictive PR group will be subjected to PR only in case of severe agitation defined by a RASS ≥ +3. Physical restraint will consist of wrist straps. In both groups, patients will receive standardized management for analgesia, sedation, delirium detection, weaning and early mobilization according to current guidelines. Concealment will be obtained using a computer-generated randomization scheme of various-sized blocks stratified by center, age (< or ≥ 65 years) and coma at the beginning of invasive mechanical ventilation (D0)) through a centralized 24h/24h internet service. Investigation blinded to group assignment is not feasible. In both arms, patients' arousal will be evaluated twice a day until day 14 with the use of RASS. Patients with a RASS of -5 or -4 will be considered comatose (and will not be assessed for delirium). Patients with a RASS score ≥ -3 will be assessed for delirium with the use of the CAM-ICU scale twice a day.

Patients in this group will be subjected to initial systematic PR, which will be re-evaluated every day every day between day 0 and day 14

Patients in this group will be subjected to PR only in case of severe agitation, defined by a RASS ≥ +3 on any given day between day 0 and day 14

number of delirium-free days during the first 14 days (D14) after randomization (D0) measured by Confusion Assessment Method for the ICU (CAM-ICU)

to assess whether a restrictive use of PR, in comparison to a systematic use, decreases delirium duration during the first 14 days (D14) after randomization (D0)

Number of coma free days during the first 14 days (D14) after randomization (D0) measured by RASS score

Median of Mobilisation capacity and rate of patients > 2 on a visual scale (SOMS) ranging from 0 (no mobilisation) to 4 (ambulation) between Day0- Day14

Rate of patients with pressor ulcer on the wrists and with other bedsores and their severity according to the National Pressure Ulcer Advisory Panel between Day0-Day14 (at least one ulcer of grade III or IV per patient)

Rate of patients with at D90 an altered cognitive capabilities defined as a MMSE (Mini Mental State Examination) ≤ 24 points

Rate of patients with a frontal syndrome defined as a FAB (Frontal Assessment Battery at Bedside) < 15 points

Rate of patients with a possible diagnosis of Post-Traumatic Stress Disorder (PTSD) defined as a R-IES (Revised-Impact of events scale) ≥ 33 points

Rate of patients with a functional disability defined as a GOS-E (Glasgow Outcome Scale - Extended) ≤ 6 points

Functional independence status (yes or no) evaluated by the FIM (functional independence measurement) scale

Inclusion criteria : Patients fulfilling all the following criteria will be eligible: Adults ≥ 18 years MV expected for at least ≥ 48 hours Invasive MV in the ICU for a duration inferior to 6 hours eligible to prescription for physical contention Exclusion criteria : Patients meeting one of the following criteria will not be considered for inclusion: Documented delirium prior to ICU admission according to the CAM-ICU History of dementia (Mini mental test < 24) Alcoholic withdrawal syndrome expected Admission for any neurological disease including post-cardiopulmonary resuscitation (including cardiac arrest, stroke, traumatic brain injury, meningoencephalitis, and status epilepticus) Serious auditory or visual disorders Unable to understand French Pregnant or lactating women SAPS II > 65 points at screening Do-not-resuscitate orders (advance directives) No affiliation to a social security regime (beneficiary or assignee) Patient or person of confidence (if present at the time of inclusion) opposing the patient's participation in research Patient already involved in another interventional clinical research whose main objective is related to delirium