Methods
Study Design and Patients
Each participating centre's Human Research Ethics Committee approved this study with either person-responsible consent or opt-out consent at the time of post-discharge follow-up (please refer to the Acknowledgements). Opt-out consent was approved at most centres because this was an observational study with no change to standard care. We recruited patients from 12 ICUs in Australia and New Zealand, including six tertiary hospitals, four metropolitan hospitals and two rural hospitals. Between August 2012 and March 2013 each unit recruited up to a maximum of 25 patients. Patients were eligible for inclusion if they were independently able to mobilise prior to the current hospital admission (this included patients who used a walking stick or gait aid to mobilise, but not patients that needed assistance from another person or a machine such as a wheelchair), had been in the ICU <72 hours, had been receiving invasive ventilation for >24 hours and had expected to stay invasively ventilated for at least the next 48 hours. Patients were excluded if they had one or more of the following: age <18 years, proven or suspected neurological impairment, inability to communicate in English, cognitive impairment prior to the ICU admission, unstable fractures or any other injury that would require specific medical bed rest orders, an ICU admission for palliative care or proven or suspected primary myopathic or neurological process associated with prolonged weakness or ICU readmission.
Study Procedures
We conducted the study in collaboration with the Australia and New Zealand Intensive Care Research Centre and the study was endorsed by the Australia and New Zealand Intensive Care Society Clinical Trials Group. We designed an online data case report form. We trained site investigators and research coordinators in all study procedures.
Measurements and Data Collection
Research coordinators at each participating site screened the patients for eligibility and sought consent as required. Research coordinators collected demographic data including age, gender and body weight, admission source, functional co-morbidities using the functional co-morbidities index, date and time of ICU and hospital admission, and date and time of first intubation at the time of enrolment into the study. We recorded the APACHE II score utilising data from the first 24 hours of admission to ICU and daily therapeutic interventions, including administration of mechanical ventilation, vasoactive agents and renal replacement therapy.
Research coordinators or site investigators collected daily data for the first 14 days of mechanical ventilation or until ICU discharge or death, whichever occurred first. Data were collected daily at 12:00 noon and included physiological information, Richmond Agitation and Sedation Scale score and maximum level of mobilisation using the ICU mobility scale. Data were also collected daily during mechanical ventilation about physiotherapy–patient interactions. Every patient was assessed by a physiotherapist to determine their ability to perform early mobilisation, as part of the standard care in each participating hospital. For patients who received early mobilisation, we collected data for all of the physiotherapy sessions on the day that the patient was seen regarding the duration of mobilisation activities, types of mobilisation activities and co-interventions (such as continuous renal replacement therapy or vasoactive agents). For patients who did not receive early mobilisation, we collected the reported barriers to mobilisation.
Early mobilisation was defined as any active exercise where the patients could assist with the activity using their own muscle strength and control that occurred while the patient was receiving invasive ventilation and was scored using the ICU mobility scale. This included the activities of rolling, bridging, sitting, standing and walking, and upper and lower limb flexion and extension, and did not preclude the patient receiving assistance from staff or equipment.
Serious adverse events were prospectively defined as a fall, unplanned extubation, cardiac arrest, loss of an invasively inserted line and new-onset atrial or ventricular tachyarrhythmia, and were recorded during mobilisation sessions. We defined adverse events that required a mobilisation session to be stopped prematurely a priori. These events were a decrease in mean arterial pressure <60 mmHg or a decrease in oxygen saturation <88% for more than 3 minutes and occurrence of a new oxygen requirement for a fraction of inspired oxygen >0.6.
Research coordinators collected ICU and hospital outcome data in all patients, including mobilisation data such as time to first mobilisation activity, time to first sit out of bed, stand and walk. We defined successful extubation as cessation of mechanical ventilation without reinitiation of ventilation within 24 hours.
In patients who were discharged from the ICU during business hours, physiotherapists assessed muscle strength using the Medical Research Council Manual Muscle Test Sum Score (MRC-SS). Full strength is a MRC-SS of 60/60 and ICUAW has been defined as a score <48/60. We recorded survival status at ICU and hospital discharge and day 90. We contacted patients by telephone at 6 months to establish their level of mobilisation using the ICU mobility scale, survival status, health-related quality of life using the EuroQoL standardised health outcome tool and status relating to undertaking paid work.
Statistical Analysis
Normally distributed variables were compared using Student's t test and reported as mean (standard deviation), while non-normally distributed variables were compared using Wilcoxon rank-sum tests and reported as median (interquartile range). Survival analysis between patients with and without ICUAW at ICU discharge was presented using a Kaplan–Meier curve with comparisons between groups performed using a log-rank test. Statistical analysis was performed using SAS version 9.3 (SAS Institute Inc., Cary, NC, USA). Two-sided P = 0.05 was considered statistically significant.