The Association between Frailty and Short-Term Outcomes in an Intensive Care Unit Rehabilitation Trial: An Exploratory Analysis

Background: Physical therapy initiated early in an ICU stay may reduce functional deficits in critically ill patients; however, the association of frailty with outcomes in those receiving early in-ICU rehabilitation is unknown. Objective: To estimate the association between frailty and 3 outcomes in patients enrolled in an ICU randomized clinical trial (RCT). Design: Exploratory secondary analyses of the CYCLE pilot RCT (NCT02377830). Setting: 7 Canadian ICUs. Participants: Previously ambulatory critically ill adults. Intervention: Participants were randomized to early in-bed cycling plus routine physiotherapy versus early routine physiotherapy alone. Measurements: Using regression analyses, we modelled the association between pre-hospital Clinical Frailty Scale (CFS) scores, Physical Function in ICU Test-scored (PFIT-s), muscle strength, and mortality at hospital discharge, adjusting for illness severity (APACHE II) and the randomized intervention. We explored the influence of imputing mean PFIT-s and strength scores for decedents, and with listwise deletion of decedents in a sensitivity analysis. Results: Of 66 patients, 2 had missing data, 2 had incomplete data, and 21 died by hospital discharge. At hospital discharge for 66 patients, frailty was not associated with PFIT-s (mean difference (MD) [95% CI]=0.20, [-2.08, 2.74]) or muscle strength (1.96, [-12.6, 16.6]). A sensitivity analysis yielded consistent results. Frailty was also not associated with hospital mortality (odds ratio 0.91, [0.28 to 2.93]). Conclusion: We found no association between pre-hospital frailty, physical function, strength, or mortality at hospital discharge in critically ill patients enrolled in an early rehabilitation trial. Larger sample sizes are needed to further explore the association of frailty with these outcomes at hospital discharge.