Should New Data on Rehabilitation Interventions in Critically Ill Patients Change Clinical Practice? Updated Meta-Analysis of Randomized Controlled Trials.

OBJECTIVES:We published a meta-analysis in March 2020 to assess the impact of rehabilitation in the ICU on clinical outcomes. Since then, 15 new randomized controlled trials (RCTs) have been published; we updated the meta-analysis to show how the recent studies have tipped the scale. DESIGN:Systematic review and meta-analysis. SETTING:An update of secondary data analysis of RCTs published between January 1998 and July 2023 performed in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines. PATIENTS:Critically ill adults. INTERVENTIONS:Cycling exercises or neuromuscular electrical stimulation (NMES) or protocolized physical rehabilitation (PPR) or functional electrical stimulation-assisted cycle ergometry (FESCE) compared with standard of care. MEASUREMENTS AND MAIN RESULTS:Days on a mechanical ventilator, length of stay in ICU and at the hospital, and mortality. We found 15 RCTs (one on cycling, eight on NMES alone, four on PPR, and two on FESCE) into which 2116 patients were randomized. The updated meta-analysis encompasses a total of 5664 patients. The exercise interventions did not influence mortality (odds ratio, 1.00 [0.87-1.14]; n = 53 RCTs) but reduced the duration of mechanical ventilation (mean difference, -1.76 d [-2.8 to -0.8 d]; n = 46) and length of stay in ICU (-1.16 d [-2.3 to 0.0 d]; n = 45). The effects on the length of mechanical ventilation and ICU stay were only significant for the PPR subgroup by a median of -1.7 days (95% CI, -3.2 to -0.2 d) and -1.9 days (95% CI, -3.5 to -0.2 d), respectively. Notably, newly published trials provided consistent results and reduced the overall heterogeneity of these results. CONCLUSIONS:None of the rehabilitation intervention strategies being studied influence mortality. Both mechanical ventilation and ICU stay were shortened by PPR, this strengthens the earlier findings as all new RCTs yielded very consistent results. However, no early rehabilitation interventions in passive patients seem to have clinical benefits. Regarding long-term functional outcomes, the results remain inconclusive.