To mobilise or not to mobilise: is that the right question?

Muscle atrophy commences early, within hours of commencement of invasive mechanical ventilation [1]. Therefore, patients who receive prolonged mechanical ventilation are at high risk of muscle atrophy, severe weakness, long-term disability, and increased mortality [2–4]. To mitigate these risks, early mobilisation has been recommended in at least seven international guidelines [5]. It is a management strategy that has had widespread support from clinicians.

Early mobilisation will purportedly reduce duration of mechanical ventilation and delirium and improve functional recovery, although the evidence from individual trials has been inconsistent [6–9]. The evidence for the international guidelines has been pooled from pilot studies, phase II trials, observational studies, and expert consensus.

Clinical practice guidelines universally recommend implementation of early mobilisation activities in the intensive care unit (ICU) [5]. Each of the guidelines has made recommendations for operationalising early mobilisation differently, reflecting the heterogeneity across included studies. Almost all of the guidelines highlight the importance of considerations of the ABCDE (awakening, breathing, coordination, delirium monitoring and management, and early mobility) bundle of care [10]. As clinicians implement mobilisation, consideration for a patient’s level of sedation and delirium is crucial, as the main barrier reported to early mobilisation is sedation [6]. Other important factors for implementation include clinician expertise and the availability of specific mobilisation equipment. Daily evaluation of a patient’s physiological status for the suitability of mobility activities and a-priori considerations for terminating a mobility session should be discussed [5, 11]. However, clinical practice guidelines should be considered according to the comprehensiveness and recency of the systematic review informing the guidelines, given the recent increase in primary research evaluating early mobilisation.

Early mobilisation is defined by an exercise intervention based on a reproducible and physiological approach. The aim is to deliver early mobilisation as soon as it is safe to do so, when patients are physiologically stable [11]. There are several difficulties with interpreting the evidence for early mobilisation:

(1) There is no clear definition of “early”. Some randomised-controlled trials have defined early mobilisation as mobilisation that commenced within 2 days of invasive mechanical ventilation, whilst others have defined it as within 5 days of admission to ICU or any time in ICU (Table 1). In the largest randomised controlled trial (RCT) to date of early mobilisation published in 2022 (N = 750), the mean duration of time from ICU admission to randomisation was 2.5 days and assessment for early active mobilisation occurred on the day of randomisation in 86% of the intervention group [6]. This study did not measure or include passive mobilisation which may have started earlier. However, in the recent study published by Patel and colleagues in 2023 (N = 200), early mobilisation (passive or active) occurred within 1.1 days [7]

Table 1.

Differences between trials of early mobilisation in baseline factors and early mobilisation delivery

Study	Schweickert Lancet 2009	Denehy Crit Care 2013	Schaller Lancet 2016	Morris JAMA 2016	Moss AJRCCM 2016	Wright Thorax 2017	TEAM NEJM 2022	Patel Lancet Resp Med 2023
N randomised (intervention/comparison)	104 (49/55)	150 (74/76)	200 (104/96)	300 (150/150)	120 (59/61)	308 (150/158)	750 (372/378)	200 (100/100)
Location(s)	United States	Australia	Austria (1), Germany (1), United States (3)	United States	United States	United Kingdom	Australia (28), New Zealand (5), United Kingdom (8), Ireland (4), Germany (3), Brazil (1)	United States
Centres	2	1	5	1	5	4	49	1
Cohort	Medical	Mixed	Surgical	Medical	Medical	Mixed	Mixed	Mixed
Enrolment dates	6/2005–10/2007	5/2007–8/2009	7/2011–11/2015	10/2009–5/2014	8/2009–10/2014	1/2012–12/2014	2/2018–11/2021	8/2011–10/2019
Age (intervention/comparison) years	57.7/54.4	61.4/60.1	66/64	55/58	61/57	60/64	60.5/59.5	57.9/54.5
Primary admission diagnosis	Sepsis 16% Acute lung injury 55%	Pneumonia 23% Cardiac 15.3%	Sepsis 1% Trauma 26%	Acute respiratory failure	ARDS 30% Pneumonia 24%	Medical 52% Surgical 48%	Sepsis 66% Trauma 4%	Acute respiratory failure 44% Sepsis 14%
APACHE II (intervention/comparison)	20/19	19/20.7	16/17	76/75 (APACHE III)	17.9/17.4	19/19	18.2/18	23/23
RASS target	Goal directed	1 to + 1	Goal directed	No sedation protocol	N/A	− 1 to + 1	Goal directed	Clinician directed
Intervention	Exercise and mobilisation at enrolment	Exercise rehabilitation and usual care	Early, goal-directed mobilisation with a facilitator to identify and address barriers	Passive range of motion, physical therapy, progressive resistance	Intensive physical therapy activities	Intensive physical rehabilitation	Sedation minimization and daily highest possible level of mobilisation	Early mobilisation at enrolment
Passive/active mobilisation at 1° level/session of the intervention	Passive range of motion	Active bed exercises	Passive range of motion	Passive range of motion	Active bed exercises	Active bed exercises	Active exercises	Passive range of motion
Comparison	Standard care, with physical and occupational therapy delivered as ordered by primary care team	Usual care	Individual centres’ practice guidelines	Usual care	Standardised physical therapy activities	Usual care	Usual care mobilisation level	Usual care when ordered by primary team or on extubation
Duration of invasive mechanical ventilation, days (intervention/comparison)	3.4/6.1	4.4/ 4.1	NA	7/NA	10/10	4/4	6/6	2.7/3.4
Time to first therapy, days (intervention/comparison)	1.5/7.4 (from intubation)	On or after day 5 of ICU admission	NA/NA	3/7 (to physical therapy)	8/8 (from initiation of mechanical ventilation)	3/3 (from enrolment; ventilated 4 days pre-randomisation)	4/7 (from ICU admission)	1.1/4.7 (from intubation)
Proportion of patients mobilised during invasive mechanical ventilation (%) (intervention/comparison)	NA	42/41	NA	NA	NA	NA	74/73	94/6
Frequency of mobilization, days per week (intervention/comparison)	Daily, to hospital discharge or return to baseline function/NR	7/7	NA	7 to hospital discharge/5	7/3 (up to 28 days)	5/5	7/5	Daily, to hospital discharge or return to baseline function/NR
Duration of mobilisation, minutes per day (intervention/comparison)	19.2/0 (whilst receiving MV)	ICU: 15–30 min/day: ward: 60 min/day; outpatient 2 × /week (all active)/NR	21/9	NA	39 (ICU 31.3; non-ICU 45.3)/21 (ICU 21; non-ICU 22)	23/13 (includes passive and active)	20.8/8.8 (active mobilisation)	NA
Median time to mobility milestones, days (intervention/comparison)	From intubation	NA	NA	NA	NA	NA	From randomisation	From intubation
Sitting	1.7/6.6						2/4	1.1/4.8
Standing	3.2/6						3/5	1.3/5
Walking	3.8/7.3						5/7	1.7/5.6

NA not available

(2) The definition and dose of “mobilisation” may be different amongst trials, for example, starting with passive exercies versus active exercises, varied use of equipment such as cycle ergometers, tilt tables or electrical muscle stimulation, or any combinations of exercise and equipment. Evidence synthesis suggests that functional exercises, such as standing, sitting, and walking, may be the most effective [5, 12].

(3) There may be a heterogeneity of treatment effect with early mobilisation where some patients respond better than others.

International clinical practice is varied. Pragmatic trials that have used “usual care” as the comparator to the intervention group may differ between trials (Table 1). Usual care comprises over half of control groups in ICU rehabilitation trials, with varied terminology, such as usual care, standard care, conventional care, and routine care. For clinicians considering the applicability of trial results in their clinical context, it is crucial to carefully review the description of both the intervention and the control groups of the study, and compare it to their own practice (Table 1). It is clear that some trials have included a usual care arm that includes minimal mobilisation in ICU, whilst other trials have included a usual care arm that includes a substantial amount of mobilisation consistent with standard practice in their ICU.

Similar to other trials of early mobilisation, the recent phase III trial (the TEAM trial) randomised 750 critically ill patients from 49 hospitals across six countries and has questioned the safety of early mobilisation compared to standard care [6]. In this trial, participants were a mixed cohort, of whom 66% had sepsis, mean body mass index was 30, 62% received vasopressors, and 22% received renal replacement therapy. Adverse events were 2.5 times more likely to occur in the intervention group compared to usual care and were more likely to reoccur in the same patients. These resolved with cessation of the early mobilisation session. Following this trial, a systematic review of the effects of early mobilisation on 6-month functional recovery was completed [13]. It included a Bayesian analysis of the data using vague priors, and reported 75% probability that early mobilisation was associated with an increase in days alive and out of hospital at 6 months, and 95% probability that it was associated with improved physical function at 6 months. However, it was also associated with 66% chance of increased adverse events and 72% chance of increased mortality at 6 months.

Overall, usual care groups have been poorly reported in the ICU rehabilitation literature. Of the studies that have reported the activities included in usual care, the content varies considerably. Activities may include passive (no patient participation, e.g., passive range of motion), active-assisted (some participation, e.g., active-assisted range of motion), or active participation. In the TEAM trial, the usual care group reported active or active-assisted mobilisation (i.e., sitting at the edge of bed or higher) for a mean duration of 9 min per day for approximately 5 days per week, as compared to 21 min, for approximately 7 days a week, in the intervention group. Similar mobility milestones were achieved in the usual care group compared to the intervention group, but not as quickly (i.e., 1–2 days later for major mobility milestones in the usual care group) in 89% of patients, including during invasive mechanical ventilation [11]. On the other hand, the study by Patel and colleagues reported an improvement in cognitive function with early mobilisation compared to standard care. Standard care in this trial included only 6/99 (6%) patients who were mobilised during mechanical ventilation.

Whilst we do not have all of the answers about the type, timing, or dose of early mobilisation to optimise patient outcomes, the implementation of early mobilisation needs to be considered on an individual basis, using the best evidence applied to a clinicians’ own setting, optimising safety and functional recovery whilst minimising risks. Future trials of early mobilisation should provide an individualised approach, and evaluate cumulative exposure to mobilisation using novel trial designs.

Author contributions

All authors contributed to the writing of the manuscript and agreed the final version.

Funding

Open Access funding enabled and organized by CAUL and its Member Institutions.

Data availability

Data included in this manuscript was available from published manuscripts. Authors should be contacted directly for further information.

Declarations

Conflicts of interest

CH was supported by an NHMRC Investigator grant and was the lead investigator on the TEAM trial published in the NEJM in 2022 and referenced in this manuscript. MK is supported by a Canada Research Chair in Critical Care Rehabilitation and Knowledge Translation. She is leading a randomised control trial of in-bed cycle ergometry with critically ill patients, and received a loan of 3 RT300 supine ergometers from Restorative Therapies (Baltimore, MD) for her research.