The feasibility of early physical activity in intensive care unit patients: a prospective observational one-center study.

BACKGROUND Prolonged immobilization may harm intensive care unit (ICU) patients, and early mobilization has been proposed to counteract that process. We describe our experience in early rehabilitation of ICU patients, and its effects on physiologic outcomes. METHODS We included all patients who stayed in our 14-bed medical ICU for > or = 7 days and received invasive mechanical ventilation for > or = 2 days. The rehabilitation program included chair-sitting, tilting-up (with arms supported or unsupported), and walking. We collected vital signs before and after each intervention. RESULTS Over a 5-month period we studied 20 patients, after a median ICU stay of 5 days. A contraindication to the intervention was present on 230 days (43%). Sedation (15%), shock (11%), and renal support (9%) were the most frequent contraindications. We obtained complete data from 275 of 424 interventions, 33% of which were performed during mechanical ventilation. The chair-sitting intervention was the most frequent (56%), followed by the tilting-up-with-arms-unsupported intervention (25%), the walking intervention (11%), and the tilting-up arms-supported intervention (8%). The chair-sitting intervention was associated with a significant (P = .03) decline in both heart rate (mean -3.5 beats/min, 95% confidence interval [CI] -6.5 to -0.4 beats/min) and respiratory rate (-1.4 breaths/min, 95% CI -2.6 to 0.1 breaths/min), whereas blood oxygen saturation (measured via pulse oximetry [S(pO(2))]) and mean arterial blood pressure did not change significantly. Heart rate and respiratory rate similarly increased with tilting-up: 14.6 beats/min, 95% CI 10.8 to 18.4 beats/min, and 5.5 breaths/min, 95% CI 3.6 to 7.3 breaths/min with arms unsupported, and 12.4 beats/min, 95% CI 7.0 to 17.9 beats/min and 2.6 breaths/min, 95% CI -0.4 to 5.7 breaths/min with arms supported). Heart rate and respiratory rate also increased with the walking intervention: 6.9 beats/min, 95% CI 2.6 to 11.1 beats/min, and 5.9 breaths/min, 95% CI 3.8 to 8.0 breaths/min. The walking intervention significantly decreased S(pO(2)). An adverse event occurred in 13 (3%) of 424 interventions, but none had harmful consequences. CONCLUSIONS Early rehabilitation is feasible and safe in patients in the ICU for longer than 1 week. The chair-sitting intervention was associated with nonsignificant oxygenation improvement. The tilting-up intervention was an effort as intense as walking.