Intensive Care Med

https://doi.org/10.1007/s00134-023-07099-4

UNDERSTANDING THE DISEASE

Physical rehabilitation in the intensive care

© 2023 Springer-Verlag GmbH Germany, part of Springer Nature

Physical rehabilitation practiced in the intensive care controlled trials (RCTs) of physical rehabilitation inter-
unit (ICU) is a complex therapeutic interaction between ventions delivered in the ICU suggested the potential of
healthcare professionals and patients, delivered in an these therapies for improving patient outcomes [5, 6].
equally complex environment. Interventions are influ- Subsequent trials, delivering physical rehabilitation at
enced by the capability, beliefs, and actions of those deliv- various stages within the ICU, following ICU, and hos-
ering and in receipt. Similarly, studies evaluating effec- pital discharge, showed mixed results, highlighting the
tiveness of rehabilitation interventions are different than methodological nuances of designing and evaluating
drug trials; despite tailoring to the physiological condi- complex rehabilitation interventions.
tion of the patient, we cannot assume that rehabilitation
can be reliably delivered as evidenced in randomized Present and future
trials [1]. In this article, we present a viewpoint regard- Inconsistency in clinical trial data has continued to chal-
ing the evolution of clinical and research practice in ICU lenge the field. A recent multi-centre RCT of early active
physical rehabilitation, and considerations for the future. mobilization versus usual care identified no difference in
the number of days alive and out of hospital at 180 days
Past post-randomization and raised concerns regarding the
The physiological, physical, and functional impairments safety of ICU-based mobilization [7]. In contrast, a sin-
resulting from prolonged immobility are significant for gle-centre RCT demonstrated that patients who received
critically ill patients. Rapid loss of skeletal muscle mass earlier mobiliation starting with passive range of motion
occurs during the first week of ICU admission [2]. How- within 1.1 days of intubation had better cognitive out-
ever until the early 2000s, immobility, often coupled with comes compared to those receiving later mobilization
deep sedation and neuromuscular blockade, was thought (within 4.7 days) [8]. As a result, there is a need to criti-
to preserve vital healing processes until such time as cally reflect on the current state of the science, with fac-
active recovery, involving physical activity, was achiev- tors outlined below key to interpretation.
able [3]. Recognition of intensive care unit-acquired
weakness, practice change towards daily sedation inter- 1. Terminology
ruption, and the emergence of post-intensive care syn-
drome describing the multi-factorial sequelae reflected Mobilization and rehabilitation activities overlap.
in survivorship from critical illness [4] was all pivotal in The terms are often used synonymously albeit nota-
evolving the culture of care for patients throughout their ble differences in the therapeutic underpinning of
recovery pathway. Landmark data from two randomised each. Mobility is “the process of moving oneself, and
of changing and maintaining postures” [9]. Mobil-
ity can be conducted by any member of the critical
*Correspondence: [email protected] care healthcare team. In contrast, rehabilitation is “a
1
School of Rehabilitation Science, Institute for Applied Health Science, set of interventions designed to optimize functioning
McMaster University, Room 403, 1400 Main Street West, Hamilton, ON L8S
1C7, Canada
and reduce disability in individuals with health con-
Full author information is available at the end of the article ditions in interaction with their environment” [10].
 Rehabilitation interventions reflect individualized cal illness and, therefore, a population most likely
goals to address patients’ needs. Rehabilitation pro- to respond to physical rehabilitation interventions.
fessionals, such as physiotherapists and occupational Baseline factors such as age, sex, severity of illness,
therapists, have a specialized skillset with distinct comorbidities, and frailty could alter recovery trajec-
expertise to evaluate specific deficits. As the field tory. These factors could be evaluated in prespeci-
matures, harmonization of terminology, descriptions, fied subgroup analyses. Biomarkers of muscle dam-
and contributions of unique healthcare professionals age and inflammation could represent important and
is required. cost-effective methods to identify patients who may
potentially develop muscle dysfunction [15]. A high
2. Reporting inflammatory state contributes to an environment of
anabolic resistance that reduces both the nutritional
Gaps exist in reporting the frequency (number of protein incorporation into skeletal muscle and the
sessions per day, per week), intensity (rate of energy normal adaptive muscle response to exercise. Arti-
expenditure), time (duration of an individual ses- ficial intelligence approaches could identify target
sion), and type (e.g., activities) [11] of ICU physical rehabilitation groups [16]. Further investigation is
rehabilitation interventions, with more pronounced required to understand how novel approaches may
deficiencies in comparison groups [12]. Consensus inform the type and timing of interventions and their
is required on the definition of adverse and serious relationship to important patient-centred outcomes
adverse events, rehabilitation-associated safety out- of physical function.
comes, and reporting strategies. Reporting of both
planned and actual protocol fidelity is essential; with- 5. Outcomes and measurement
out detail on protocol implementation and adherence
success, it is impossible to discern whether trial find- Clinical trials are considered “positive” if the pri-
ings reflect true failure of intervention effectiveness, mary outcome demonstrates a statistically sig-
or alternatively, failure to adequately implement the nificant difference between intervention and com-
intervention [13]. parison groups. To understand the effectiveness of
ICU-based interventions, we need primary outcome
3. Usual care measures that address the mechanism of action of
the intervention and are acquired proximal to inter-
‘Usual care’ is the most common comparator group vention delivery. Some “positive” physical rehabili-
in ICU rehabilitation trials [1]. The timing and con- tation trials measured outcomes such as physical
tent of usual care relative to the intervention could function at ICU or hospital discharge, with a direct
enhance or dampen the intervention effect. However, temporal relationship to the ICU-based intervention
few studies systematically document the timing or [1]. In contrast, studies demonstrating no difference
type of usual care in ICU or on the ward. A recent between groups measured primary outcomes after
systematic review reported how using a basic defini- hospital discharge, out to and beyond 12 months. In
tion of usual care dose impacted key outcomes [1]. If these situations, where the primary outcome is meas-
usual care occurred less than 5 days/week, the effect ured more distal to the intervention, there is risk of
of rehabilitation interventions was amplified, with a confounding factors potentially diluting a treatment
decrease in duration of mechanical ventilation (MV) effect. To understand effectiveness of rehabilitation
of 1.6 days, ICU length of stay (LOS) 1.87 days, and interventions across studies, we need consistency
hospital LOS 2.4 days [1]. In contrast, if usual care in outcome measures and timepoints across trials,
was 5 days/week or more, there was no difference as recommended by core outcome sets [17]. Finally,
in MV duration, and smaller differences in ICU and consideration of patients’ perspectives, reflected
hospital LOS [1]. Future studies could also consider in patient-reported outcomes and opportunities
more sophisticated dose comparisons, as seen in an for longer term follow-up is warranted [18].
ongoing adaptive trial in acute stroke [14].
As further ICU rehabilitation evidence becomes
4. Determining target responders available, there is an onus on clinicians to critically
appraise trials to accurately interpret findings and rel-
To date, inclusion criteria into rehabilitation tri- evance to their own practice/settings. Table 1 reflects
als have not consistently identified patients most at the aforementioned items, and compares and contrasts
risk of developing the detrimental sequelae of criti- the current state of ICU rehabilitation with areas of
Table 1 Key elements of enhanced critical appraisal
Item Present state Future considerations Clinician reflections for research applicability

Population Broad eligibility criteria Recovery trajectories How similar is the study population to my patients?
Patient stratification
Identification of target responder phenotypes
Intervention Heterogeneous interventions Reporting planned protocol, actual protocol fidelity and Is there sufficient information reported about the interven-
Limited information about intervention parameters adherence, and reflect on success of protocol implemen- tion so I can reproduce it in my own setting?
tation Did the authors deliver the intervention as intended (e.g.,
Per-protocol analysis to understand relationship between frequency, intensity, type, time, total dose, etc.)?
intervention safety events and ­mortalitya
Comparison Usual care not consistently captured across jurisdictions Better characterization of usual care by frequency, intensity, Is usual care sufficiently reported and is it similar to my own
type, and time setting?
Studies that compare different well-defined rehabilitation Is the time to start similar to the intervention?
strategies
Outcomes Primary outcomes varied across multiple time points, some Use of core outcome sets to inform consistent measures What was the primary outcome, when was it measured, and
proximal, and others distal to the end of the intervention and timepoints at what timepoint in relation to intervention delivery?
Safety outcomes inconsistently defined and reported Standardized safety measures in intervention and compari- Is it reasonable to expect an intervention effect at the time
son groups of primary outcome measurement, or could other factors
influence the outcome?
Study design Typically single-centre studies, primarily academic ICUs, Is there sufficient equipoise to randomize patients to a Are the ICUs participating in the study similar to mine (e.g.,
reflecting the nature of evolving science comparison group of no physical rehabilitation? healthcare provider expertise, equipment)?
How do we identify the right dosage to evaluate?
More international multi-centre studies including academic
and community ICUs
This table describes the present state and future considerations for ICU-based physical rehabilitation interventions by population, intervention, comparision, outcomes, and study designs. By each item, we suggest
enhanced critical appraisal reflections for clinicians to apply research for their own settings
a
An intention to treat analysis is the preferred method to analyze RCT efficacy, however to understand the relationship between rehabilitation, safety events and mortality, a per-protocol analysis is more relevant
future study. In closing, we urge current researchers Scruth E, Spill GR, Storey CP, Render M, Votto J, Harvey MA (2012) Improv-
ing long-term outcomes after discharge from intensive care unit: report
to build upon previous work to enhance the transpar- from a stakeholders’ conference. Crit Care Med 40:502–509
ency and reporting of their complex interventions. 5. Schweickert WD, Pohlman MC, Pohlman AS, Nigos C, Pawlik AJ, Esbrook
We encourage the critical care community to care- CL, Spears L, Miller M, Franczyk M, Deprizio D, Schmidt GA, Bowman
A, Barr R, McCallister KE, Hall JB, Kress JP (2009) Early physical and
fully consider lessons from the past and present before occupational therapy in mechanically ventilated, critically ill patients: a
designing future ICU rehabilitation trials. randomised controlled trial. Lancet 373:1874–1882
6. Burtin C, Clerckx B, Robbeets C, Ferdinande P, Langer D, Troosters T,
Hermans G, Decramer M, Gosselink R (2009) Early exercise in critically
Author details ill patients enhances short-term functional recovery. Crit Care Med
1
School of Rehabilitation Science, Institute for Applied Health Science, 37:2499–2505
McMaster University, Room 403, 1400 Main Street West, Hamilton, ON L8S 1C7, 7. Hodgson CL, Bailey M, Bellomo R, Brickell K, Broadley T, Buhr H, Gabbe
Canada. 2 Physiotherapy Department, Research Institute of St. Joe’s Hamilton, BJ, Gould DW, Harrold M, Higgins AM, Hurford S, Iwashyna TJ, Serpa Neto
50 Charlton Avenue E, Hamilton, ON L8N 4A6, Canada. 3 Physiotherapy A, Nichol AD, Presneill JJ, Schaller SJ, Sivasuthan J, Tipping CJ, Webb S,
Department, Division of Allied Health, Austin Health, Studley Road, Melbourne, Young PJ (2022) Early active mobilization during mechanical ventilation
VIC, Australia. 4 Wellcome‑Wolfson Institute for Experimental Medicine, in the ICU. N Engl J Med 387:1747–1758
Queen’s University Belfast, 97 Lisburn Road, Belfast BT9 7BL, UK. 8. Patel BK, Wolfe KS, Patel SB, Dugan KC, Esbrook CL, Pawlik AJ, Stulberg M,
Kemple C, Teele M, Zeleny E, Hedeker D, Pohlman AS, Arora VM, Hall JB,
Funding Kress JP (2023) Effect of early mobilisation on long-term cognitive impair-
MK is funded by a Canada Research Chair in Critical Care Rehabilitation and ment in critical illness in the USA: a randomised controlled trial. Lancet
Knowledge Translation. She currently holds grants from the Canadian Insti- Respir Med. https://​doi.​org/​10.​1016/​S2213-​2600(22)​00489-1
tutes of Health Research, Canada Foundation for Innovation, and the Ontario 9. Bussmann JB, Stam HJ (1998) Techniques for measurement and assess-
Research Fund. She leads CYCLE, an international, multi-centre randomized ment of mobility in rehabilitation: a theoretical approach. Clin Rehabil
clinical trial of early in-bed cycling. Restorative Therapies (Baltimore, MD, USA) 12:455–464
loaned Dr. Kho 3 RT300 in-bed cycle ergometers to conduct her research. BC 10. World Health Organization (2023) Rehabilitation. World Health Organiza-
currently receives project grant funding from the National Institute for Health tion. https://​www.​who.​int/​news-​room/​fact-​sheets/​detail/​rehab​ilita​tion
and Care Research (UK). 11. Barisic A, Leatherdale ST, Kreiger N (2011) Importance of frequency, inten-
sity, time and type (FITT) in physical activity assessment for epidemiologi-
Data availability cal research. Can J Public Health 102:174–175
Not applicable. 12. Reid JC, Unger J, McCaskell D, Childerhose L, Zorko DJ, Kho ME (2018)
Physical rehabilitation interventions in the intensive care unit: a scoping
Declarations review of 117 studies. J Intensive Care 6:80
13. Borrelli B, Sepinwall D, Ernst D, Bellg AJ, Czajkowski S, Breger R, DeFranc-
Conflicts of interest esco C, Levesque C, Sharp DL, Ogedegbe G, Resnick B, Orwig D (2005)
The authors have no conflicts of interest to declare. A new tool to assess treatment fidelity and evaluation of treatment
fidelity across 10 years of health behavior research. J Consult Clin Psychol
73:852–860
Publisher’s Note 14. Bernhardt J, Churilov L, Dewey H, Donnan G, Ellery F, English C, Gao
Springer Nature remains neutral with regard to jurisdictional claims in pub- L, Hayward K, Horgan F, Indredavik B, Johns H, Langhorne P, Lindley R,
lished maps and institutional affiliations. Martins S, Ali Katijjahbe M, Middleton S, Moodie M, Pandian J, Parsons B,
Robinson T, Srikanth V, Thijs V (2023) A phase III, multi-arm multi-stage
Received: 20 February 2023 Accepted: 8 May 2023 covariate-adjusted response-adaptive randomized trial to determine
optimal early mobility training after stroke (AVERT DOSE). Int J Stroke.
https://​doi.​org/​10.​1177/​17474​93022​11422​07
15. Haines RW, Zolfaghari P, Wan Y, Pearse RM, Puthucheary Z, Prowle JR
(2019) Elevated urea-to-creatinine ratio provides a biochemical signature
References of muscle catabolism and persistent critical illness after major trauma.
1. Wang YT, Lang JK, Haines KJ, Skinner EH, Haines TP (2021) Physical reha- Intensive Care Med 45:1718–1731
bilitation in the ICU: a systematic review and meta-analysis. Crit Care Med 16. Fuest KE, Ulm B, Daum N, Lindholz M, Lorenz M, Blobner K, Langer N,
50(3):375–388 Hodgson C, Herridge M, Blobner M, Schaller SJ (2023) Clustering of criti-
2. Fazzini B, Märkl T, Costas C, Blobner M, Schaller SJ, Prowle J, Puthucheary cally ill patients using an individualized learning approach enables dose
Z, Wackerhage H (2023) The rate and assessment of muscle wasting dur- optimization of mobilization in the ICU. Crit Care 27:1
ing critical illness: a systematic review and meta-analysis. Crit Care 27:2 17. Connolly B, Denehy L, Hart N, Pattison N, Williamson P, Blackwood B
3. Brower RG (2009) Consequences of bed rest. Crit Care Med 37:S422-428 (2018) Physical rehabilitation core outcomes in critical illness (PRACTICE):
4. Needham DM, Davidson J, Cohen H, Hopkins RO, Weinert C, Wunsch H, protocol for development of a core outcome set. Trials 19:294
Zawistowski C, Bemis-Dougherty A, Berney SC, Bienvenu OJ, Brady SL, 18. Herridge MS, Azoulay É (2023) Outcomes after critical illness. N Engl J
Brodsky MB, Denehy L, Elliott D, Flatley C, Harabin AL, Jones C, Louis D, Med 388:913–924
Meltzer W, Muldoon SR, Palmer JB, Perme C, Robinson M, Schmidt DM,