We aimed to quantify improvements in sensation, strength, balance, walking and dexterity over time and determine participation in daily life 1.5 years after critical illness onset. Furthermore, the predictive capacity of sensorimotor assessments for participation was analysed.
Critical illness survivors, mechanically ventilated ≥5 days, and with weakness at admission to neurorehabilitation were followed within a prospective single-centre cohort study in Germany. A wide range of sensorimotor assessments were performed, including sensory testing (superficial, vibration, proprioception), the Medical Research Council Scoring System, Five-Times-Sit-to-Stand-Test, Functional Ambulation Categories, Mini Balance Evaluation Systems Test (Mini-BESTest), Functional Reach Test and Box-and-Block-Test. Measurements were performed at admission and discharge from inpatient neurorehabilitation, and 1.5 years after critical illness onset. Time effects were analysed using a mixed-effects model. Participation was assessed with the Reintegration to Normal Living Index 1.5 years after critical illness onset. All sensorimotor assessments at the start of rehabilitation were included in a linear regression analysis with variable selection to identify predictors for participation 1.5 years after illness onset. The model was compared to an extended regression model containing demographic variables and factors previously shown to be associated with participation.
A total of 163 critical illness survivors (mean age 61.7±13.3 years, 56 females) were included, admitted to intensive care mainly because of Covid-19, as well as for cardiac and pulmonary disease (together >50%). All sensorimotor outcomes improved over time, except sensation. However, deficits remained after rehabilitation and in the long term. Good participation (≥75%) was achieved by 61.4% of survivors 1.5 years after critical illness onset (median Reintegration to Living Index rate=81.8%). To predict participation only with the sensorimotor outcomes at rehabilitation start, the Mini-BESTest and Functional Reach Test were included in the selected model (R²=0.055, p=0.011). An extended regression analysis to predict participation resulted in a model (R²=0.21, p0.001) with the variables sex, duration of mechanical ventilation, Covid-19 status, cognitive function, Mini-BESTest and Functional Reach Test.
We observed significant improvements in sensorimotor functions with lingering deficits in sensation, strength, balance, dexterity and participation. Balance outcomes at the beginning of rehabilitation demonstrated a modest yet significant predictive value for future social participation.
With the detailed understanding of sensorimotor recovery and long-term participation, health care providers are better equipped to offer patients and caregivers realistic expectations, tailored treatment approaches and recommendations to enhance participation in daily life after rehabilitation discharge. Multicentre studies and clinical research in this population is needed to investigate the connection of balance and participation further.
recovery of function
social participation
