Leg Cycle Ergometry in Critically Ill Patients - An Updated Systematic Review and Meta-Analysis.

Abstract

Background: Cycle ergometry is a rehabilitation strategy used in the intensive care unit (ICU) which may help mitigate post-ICU impairments. We aimed to systematically review and summarize evidence examining the efficacy and safety of cycle ergometry in the ICU.

Methods: We included randomized controlled trials (RCTs) of critically ill adults with any diagnosis admitted to the ICU for >24 hours, comparing cycling interventions to control (no cycling). The primary outcome was physical function, using a hierarchical approach to standardize this outcome across trials. We performed random-effects meta-analyses and assessed the certainty of effect estimates using the Grading of Recommendations Assessment, Development, and Evaluation approach.

Results: We included 33 RCTs that enrolled 3274 patients. Cycling may improve physical function at ICU discharge (12 RCTs, 1291 patients, standardized mean difference [SMD], 0.33 [95% confidence interval (CI), 0.05 to 0.62], low certainty) and posthospital discharge (8 RCTs, 865 patients, SMD, 0.23, [95% CI, 0.04 to 0.42], low certainty). Cycling may decrease ICU length of stay (29 RCTs, 2575 patients, mean difference [MD], 1.06 days fewer [95% CI, 0.33 to 1.80 days fewer], low certainty) and probably decreases hospital length of stay (22 RCTs, 2060 patients, MD, 1.48 days fewer [95% CI, 0.47 to 2.49 days fewer], moderate certainty). Cycling may have no effect on ICU mortality (17 RCTs, 2039 patients, risk ratio, 12 fewer deaths per 1000 [95% CI, 43 fewer to 23 more], low certainty). The pooled rate of adverse events in the intervention group was 1% (11 RCTs, 4623 sessions, [95% CI, 0 to 2%], low certainty) and in the comparison group, 2% (6 RCTs, 3365 sessions, [95% CI, 0 to 5%], low certainty).

Conclusions: In this meta-analysis, we found that cycling with critically ill patients may improve physical function at ICU discharge and after hospital discharge, may reduce ICU length of stay, and probably reduces hospital length of stay, with no effect on other outcomes including mortality. We observed low to very low certainty of evidence for all but one outcome of interest. Adverse events were uncommon. (PROSPERO number, CRD 42018092132.).