Findings from the Restrictive Use of Restraints and Delirium Duration in ICU (R2D2-ICU) trial were presented at the 45th ISICEM in Brussels.
The trial investigated whether a restrictive (low-use) versus liberal (high-use) wrist-strap physical restraint strategy improves outcomes in critically ill adults receiving invasive mechanical ventilation. Physical restraints are widely used in ICUs, primarily to prevent self-extubation and accidental device removal, yet their benefits remain uncertain, and concerns persist regarding potential harms, including increased agitation, sedation requirements, delirium, and long-term cognitive impairment. This study aimed to determine whether reducing restraint use could decrease the duration of coma or delirium and improve patient outcomes.
The trial was conducted in 10 ICUs in France between January 2021 and January 2024. A total of 405 adult patients were enrolled, all of whom had initiated invasive mechanical ventilation within the previous six hours and were expected to require ventilation for at least 48 hours. Patients with pre-existing delirium, dementia, neurological conditions, or other specified exclusions were not included. Participants were randomised in a 1:1 ratio to either a low-use restraint strategy (restraints applied only in cases of severe agitation, defined as a Richmond Agitation-Sedation Scale [RASS] score ≥3) or a high-use strategy (routine application of restraints with daily reassessment).
The primary outcome was the number of days alive without coma or delirium during the first 14 days after randomisation. Secondary outcomes included delirium incidence and duration, sedation and analgesia exposure, self-extubation, unplanned device removal, pressure ulcers, ventilator-free days, ICU and hospital length of stay, mortality at 90 days, and long-term functional, cognitive, and psychological outcomes.
Of the 405 randomised patients, 396 were included in the primary analysis. Baseline characteristics were well balanced between groups, with a median age of approximately 65 years, a predominance of male patients, and comparable severity of illness as measured by Sequential Organ Failure Assessment scores. Most patients were comatose at enrolment and had been admitted for medical reasons, predominantly acute respiratory failure.
In the low-use group, restraints were applied in 36.2% of patients on any given day, with a median duration of 0 hours, whereas in the high-use group, restraints were applied to all patients, with a median duration of approximately 16.8 hours per day. However, protocol nonadherence occurred in both groups, potentially influencing outcomes.
The primary outcome showed no statistically significant difference between strategies. Patients in the low-use group had a mean of 6.67 days alive without coma or delirium compared with 6.30 days in the high-use group. Thus, reducing restraint use did not improve the primary endpoint.
Secondary outcomes were similarly comparable between groups. The incidence of delirium was high in both groups (62.2% in the low-use group versus 67.5% in the high-use group), with no significant difference. Duration of delirium, days alive without coma, and levels of agitation were also similar. Importantly, exposure to sedatives, analgesics, and antipsychotic medications did not differ significantly, indicating that reduced restraint use did not necessitate increased pharmacological sedation.
Safety outcomes were likewise comparable. Self-extubation occurred in 9.2% of patients in the low-use group and 8.5% in the high-use group, while unplanned device removal and pressure ulcer rates were low and similar between groups. Mortality at 90 days was also not significantly different (37.2% vs 41.0%). Additionally, no meaningful differences were observed in functional independence, cognitive performance, or post-traumatic stress symptoms at 90-day follow-up.
The authors highlight that this is the first multicentre randomised trial directly comparing restraint strategies in mechanically ventilated ICU patients, providing high-quality evidence in an area previously guided largely by expert opinion. The lack of benefit suggests that delirium and coma in critically ill patients may be driven more by underlying illness and systemic factors than by restraint use alone.
Despite the neutral findings, the study suggests that a low-use restraint strategy can be safely implemented within a structured ICU care framework without increasing adverse events or sedative exposure. However, potential benefits in unmeasured domains, such as patient experience or psychological burden, cannot be excluded.
In conclusion, among mechanically ventilated ICU patients, a restrictive, low-use wrist restraint strategy did not reduce the duration of delirium or coma compared with a liberal, high-use approach. Both strategies resulted in similar clinical, safety, and long-term outcomes, suggesting that reducing restraint use alone is insufficient to improve neurological outcomes in this population.
Source: JAMA
Image Credit: iStock
