[Efficacy and safety of early physical therapy for acute gastrointestinal injury during mechanical ventilation in patients with sepsis: a randomized controlled pilot trial].

Objective: To investigate the therapeutic effect and safety of early physical therapy for acute gastrointestinal injury (AGI) in septic patients receiving mechanical ventilation.
Methods: A randomized controlled trial was conducted in the ICU of a tertiary teaching hospital from May, 2017 to March, 2018. The patients diagnosed with sepsis complicated by AGI during mechanical ventilation were recruited and block-randomized into intervention group and control group. Both groups received standard therapy of sepsis, and the patients in the intervention group also received physical therapy as soon as they were hemodynamically stable. The outcome measures included the recovery of AGI, ICU mortality, duration and outcomes of mechanical ventilation and the length of ICU stay.
Results: A total of 60 patients were initially included, and 34 of them completed the study, including 16 in the intervention group and 18 in the control group. After physical rehabilitation, the number of patients with a cure of AGI did not significantly differ between the two group ( P > 0.05). Nonetheless, the reduction of AGI scores after the treatments differed significantly between the intervention group and the control group (-1.9±2.1 vs 0.9± 1.6, P < 0.05). No significant differences were found between the two groups in ICU mortality, duration and outcomes of mechanical ventilation, or the length of ICU stay ( P > 0.05). In the intervention group, the incidence of exercise-related adverse events was 3.33%, and severe organ injury or death occurred in none of patients.
Conclusions: Early rehabilitation therapy does not reduce the incidence of AGI but can lower AGI scores and alleviate gastrointestinal symptoms in patients with sepsis during mechanical ventilation. The results still await further verification by welldesigned multicenter clinical trials with large sample sizes.