Robotic-Assisted In-Bed Mobilization in Ventilated ICU Patients With COVID-19: An Interventional, Randomized, Controlled Pilot Study (ROBEM II Study)*

OBJECTIVES: The COVID-19 pandemic significantly impacted global healthcare systems, particularly in managing critically ill mechanically ventilated patients. This study aims to assess the feasibility of robotic-assisted mobilization in COVID-19 patients. DESIGN: Randomized controlled pilot study. SETTING: Four COVID-19 specialized ICUs at Charité-Universitätsmedizin Berlin (March 2021 to February 2022). PATIENTS: Twenty critically ill COVID-19 patients expected to require greater than 24 hours of ventilation. INTERVENTIONS: A 5-day intervention phase with bid robotic-assisted mobilization greater than or equal to 20 minutes and follow-up at day 180, compared with standard care. MEASUREMENTS AND MAIN RESULTS: Intervention sessions were conducted in 98.9% according to protocol, with one session missing due to staff shortage. Primary outcome was the mobilization level measured with the ICU Mobility Scale (IMS) and Surgical ICU Optimal Mobilization Score (SOMS), assessed until day 5 or extubation. Safety events were recorded during mobilization. The median IMS and SOMS were 0 (0-0.16) and 1 (1-1.03) in the intervention group, and 0 (0-0.15) ( p = 0.77) and 0.8 (0.65-1.20) ( p = 0.08) in the standard care group, respectively. Significant secondary outcomes included average number of mobilization sessions (intervention: 8.5 [7.75-10] vs. standard care: 4.5 [3.5-5]; p = 0.001), total mobilization time (intervention: 232.5 min [187.25-266.5 min] vs. standard care: 147.5 min [107.5-167.5 min]; p = 0.011), and healthcare providers per session (intervention: 2 [2-2] vs. standard care: 1 [1-1.4]; p = 0.001) during intervention. Four safety events (hypertension and agitation, n = 2 each) in the intervention group and none in the standard care group were reported. CONCLUSIONS: Robotic-assisted mobilization in mechanically ventilated COVID-19 patients appears to be safe and feasible.