Journal Pre-proof

Early Rehabilitation Reduces the Likelihood of Developing Intensive Care

Unit-Acquired Weakness: A Systematic Review and Meta-Analysis

David E. Anekwe, Sharmistha Biswas, André Bussières, Jadranka

Spahija

PII: S0031-9406(19)30128-2
DOI: https://doi.org/10.1016/j.physio.2019.12.004
Reference: PHYST 1156

To appear in: Physiotherapy

Please cite this article as: Anekwe DE, Biswas S, Bussières A, Spahija J, Early Rehabilitation
Reduces the Likelihood of Developing Intensive Care Unit-Acquired Weakness: A Systematic
Review and Meta-Analysis, Physiotherapy (2019),
doi: https://doi.org/10.1016/j.physio.2019.12.004

This is a PDF file of an article that has undergone enhancements after acceptance, such as
the addition of a cover page and metadata, and formatting for readability, but it is not yet the
definitive version of record. This version will undergo additional copyediting, typesetting and
review before it is published in its final form, but we are providing this version to give early
visibility of the article. Please note that, during the production process, errors may be
discovered which could affect the content, and all legal disclaimers that apply to the journal
pertain.

© 2019 Published by Elsevier.

Title: Early Rehabilitation Reduces the Likelihood of Developing Intensive Care Unit-
Acquired Weakness: A Systematic Review and Meta-Analysis

Authors: David E. Anekwe PhD.,1,2,3 Sharmistha Biswas, MBBS., MSc,4,5 André Bussières

PhD.,1,3 Jadranka Spahija PhD.1,2,3

1. School of Physical and Occupational Therapy, McGill University, Montreal, Quebec,

Canada

2. Research Center, CIUSSS du Nord-de-l'Ile-de-Montréal, Sacré-Coeur Hospital,

Université de Montréal, 5400 Boul. Gouin Ouest, Montréal, Quebec, H4J 1C5,

of
Canada.

ro
3. Center for Interdisciplinary Research in Rehabilitation in Montreal, CISS du Nord-de-

l'Île-de-Montréal, Jewish Rehabilitation Hospital, 3205, Place Alton-Goldbloom,

Laval, Quebec, H7V 1J1, Canada.

-p
re
4. Respiratory Epidemiology and Clinical Research Unit, Research Institute of the

McGill University Health Centre, Montreal, Quebec, Canada.

lP

5. Epidemiology, Biostatistics and Occupational Health, McGill University, Montreal,

Canada.
na

Corresponding Author: Jadranka Spahija PhD

CIUSSS du Nord-de-l'Ile-de-Montréal Hôpital du Sacré-Cœur de Montréal
Axe de recherche en maladies chroniques 5400 boul. Gouin Ouest,
ur

Montréal, Quebec, H4J 1C5, Canada

Telephone: (514) 338-2222 ext.3654
Jo

Fax : 514-739-7357
jadranka.spahija@mcgill.ca

Word Count: 2979 words (Introduction, Method, Results, Discussion)

ICU Rehabilitation and ICU-Acquired Weakness 17/12/2019

Page 1 of 24
ABSTRACT

Background: Intensive care unit-acquired weakness (ICUAW) is associated with significant

impairments in body structure and function, activity limitation, and participation restriction.

The etiology and management of ICUAW remain uncertain.

Objective: To estimate the extent to which early rehabilitation interventions (early mobilization

[EM] and/or neuromuscular electrical stimulation [NMES]) compared to usual care reduce the

incidence of ICUAW in critically ill patients.

of
Data sources: We searched MEDLINE, EMBASE, CINAHL, Cochrane Central and

ro
Physiotherapy Evidence Database databases from inception to May 1st, 2017.

-p
Eligibility criteria: Randomized controlled trials of EM and/or NMES interventions in

critically ill adults.

re
Data extraction and data synthesis: Data on the incidence of ICUAW and secondary
lP

outcomes were extracted. Both odds and risk ratios for ICUAW were pooled using the random-

effects model.
na

Results: We identified 1421 reports after duplicate removal. Nine studies including 841

patients (419 intervention and 422 usual care) were included in the final analysis. The

interventions involved EM in five trials, NMES in three trials, and both EM and NMES in one
ur

trial. Early rehabilitation decreased the likelihood of developing ICUAW: odds ratio of 0.63
Jo

(95% CI: 0.43-0.92) in the screened population, and 0.71 (95% CI: 0.53-0.95) in the randomized

population.

Conclusion, implications of key findings: Early rehabilitation was associated with a decreased

likelihood of developing ICUAW. Our findings support early rehabilitation in the ICU. While

ICU Rehabilitation and ICU-Acquired Weakness 17/12/2019

Page 2 of 24
results were consistent in both the screened and randomized populations, the wide confidence

intervals suggest that well-conducted trials are needed to validate our findings.

Keywords: Early mobilization, Neuromuscular Electrical Stimulation, Intensive

Care Unit Acquired Weakness, Mortality, Mechanical Ventilator

Duration, Discharge Location

of
ro
INTRODUCTION

-p
Intensive care unit (ICU) survivors often develop muscle weakness which is unrelated to the

primary pathology for ICU admission.[1] This weakness may progress to a clinical syndrome
re
known as ICU-Acquired Weakness (ICUAW), a “clinically detected weakness in critically ill

patients in whom there is no plausible etiology other than critical illness”.[2] This syndrome
lP

includes critical illness polyneuropathy, myopathy, and neuromyopathy.[2, 3]

ICUAW is associated with significant impairments in body structure and function,[4, 5] activity
na

limitations,[4-6] and participation restrictions.[4] It has a long-lasting impact,[4-6] persisting

for months or years following ICU discharge with a resultant decline in quality of life for ICU
ur

survivors[5, 7] and places a high burden on the healthcare system.[8] There is currently no
Jo

effective treatment for ICUAW.[9, 10] Although it is often recommended that patients with

ICUAW be referred for rehabilitation, a 2015 Cochrane review[11] failed to identify studies to

determine whether physical rehabilitation improve activities of daily living, muscle strength

and quality of life in these ICU survivors. Thus, preventing the development of ICUAW through

the control of associated risk factors [9] such as immobility[12] is key. Immobility predisposes

ICU Rehabilitation and ICU-Acquired Weakness 17/12/2019

Page 3 of 24
to critical illness myopathy through protein loss resulting from altered protein metabolism[13]

and to critical illness polyneuropathy through hyperglycemia as a result of insulin

resistance.[13] Therefore, reducing the level of immobility during ICU admission may decrease

the risk of developing ICUAW.[9, 10]

Early mobilization (EM) reduces the duration of immobility, while neuromuscular electrical

stimulation (NMES) reduces muscle atrophy[14] which is involved in the pathophysiological

process of ICUAW.[15, 16] Both interventions are sometimes used together as part of early

rehabilitation intervention in the ICU, however, few robust studies have evaluated the

of
effectiveness of EM and NMES to reduce the likelihood of developing ICUAW. To date, few

ro
systematic reviews[17-19] have evaluated the effectiveness of EM and NMES to reduce the

risk of ICUAW, and these have included only two[17, 19] and three[18] RCTs. Additional
-p
trials are currently available. [20-24] Moreover, the meta-analysis in last two systematic

reviews[18, 19] analyzed patients with missing data (in some of the included RCTs) as not
re
having ICUAW. In these studies, ICUAW was assessed with the MRC sum score, which is
lP

prone to having missing data in very weak and uncooperative patients. Given this context, it

seems more appropriate to conservatively err on the side of inclusion and impute the worst
na

values (= ICUAW) for patients who were not evaluable. Furthermore, the forest plot on the

review[18] also had errors in the incidence of ICUAW for both the intervention and control
ur

groups for the study by Routsi et al.[25] The few trials included in two previous reviews also

limited the exploration of sources of heterogeneity. This review is, therefore, necessary to revise
Jo

and update the knowledge currently available and the specific research questions was: to what

extent do early rehabilitation interventions of EM and NMES, compared to usual care, reduce

the incidence of ICUAW among patients in the ICU, and alter other outcomes (length of time

on mechanical ventilator, discharge location, ICU and hospital length of stay, and acute

mortality) that may be associated ICUAW?

ICU Rehabilitation and ICU-Acquired Weakness 17/12/2019

Page 4 of 24
METHOD

Identification and selection of studies

We searched MEDLINE, EMBASE, CINAHL, Cochrane Central Register of Controlled Trials

and Physiotherapy Evidence Database databases, from inception to May 1st, 2017, without

language restriction (Supplementary Digital Content [SDC] I). Studies were included if they:

(i) were done in the ICU, (ii) were RCTs, (iii) involved adult participants, (iv) evaluated the

effect of EM or NMES interventions, and (v) reported the incidence of ICUAW or assessed

of
muscle strength using the Medical Research Council Scale (MRC). Studies that evaluated

ro
rehabilitation interventions in patients already diagnosed with ICUAW were excluded.

Two reviewers independently screened the titles and abstracts of retrieved articles, and the full
-p
texts of potentially eligible articles were obtained and further assessed for final inclusion.
re
Reviewers formally met at each step to reach a consensus, and a senior author resolved

differences if needed.
lP

Box 1. Inclusion criteria

na
ur
Jo

ICU Rehabilitation and ICU-Acquired Weakness 17/12/2019

Page 5 of 24
 Design

 Randomized controlled trial

Participants

 Adults
 Critically ill participants patients not already diagnosed with intensive care unit
acquired weakness (ICUAW)
Intervention

 Evaluated the effect of early rehabilitation interventions of early mobilization

(EM) and/or neuromuscular electrical stimulation (NMES) interventions
Outcome measures

of
 Reported the incidence of ICUAW or assessed muscle strength using the
Medical Research Council Scale (MRC)
Comparisons

ro
 Usual care

-p
Data extraction and assessment of characteristics of studies
re
Two reviewers participated in the data extraction and quality assessment of the individual

studies using the Cochrane Risk of Bias Tool.[26] Population characteristics, treatment
lP

intervention(s), control/comparators, and outcomes were described for each included study.

Quality assessment was performed in duplicate by the reviewers; items were categorized as
na

‘Low, High or Unclear’ based on the criteria shown in the SDC II. All disagreements were

tracked and resolved by discussions and consensus or by a senior author if needed. Authors of
ur

the primary studies were contacted to request any important missing information.

Data analysis
Jo

The primary outcome was the incidence of ICUAW measured at any time point after the

initiation of the intervention. The secondary outcomes were length of time on mechanical

ventilator (ventilator-free days and duration of mechanical ventilation), discharge location, ICU

and hospital LOS, and acute mortality (defined as death in the ICU or hospital). For binary

ICU Rehabilitation and ICU-Acquired Weakness 17/12/2019

Page 6 of 24
outcomes, we extracted the number of events and total number in each group. For continuous

outcomes, we extracted the mean (standard deviation) or the median (range) values.

Data were analyzed using Stata 14 (StataCorp., 2015). We expected some difference in the true

effect in the population (heterogeneity) given the differences in the patients studied and the

types of rehabilitation interventions used in the studies. We, therefore, used the random-effect

model to pool the effect sizes from the different studies in a meta-analysis and performed a

sensitivity analysis using the fixed effect model.[27] Only dichotomous data were pooled

together. A priori, we planned to report only risk ratios (RR), but during our review, we

of
discovered that the true control group risk (without exposure to the intervention) was unknown

ro
as the control group in six of the studies had EM as part of standard care. Since, RR are bound

by the control group risk, while odds ratios (OR) are not,[28, 29] we judged that OR is
-p
mathematically more appropriate for this meta-analysis.[28, 30, 31] We, therefore, reported

both measures (RR and the OR) but with a focus on OR (95% confidence intervals). For all the
re
studies, we obtained data for patients who were evaluated for ICUAW (the screened population)
lP

and the total randomized populations, and calculated the OR for both the screened population

and randomized population (with the imputation of worst values for patients who had missing

data [intention-to-treat analysis] – as was done in a previous Cochrane review[17]). Statistical

na

heterogeneity was assessed with the I2 statistic, while clinical heterogeneity was explored with

subgroup analysis based on (i) short vs long ICU LOS (with short ICU LOS defined as ≤ 7 days
ur

in both the intervention and control group), (ii) the timing of intervention, (iii) the type of
Jo

rehabilitation intervention (EM or NMES), (iv) different time points of ICUAW assessment

(ICU awakening, 7th day post-awakening, ICU discharge, hospital discharge). The latest

measuring timepoint was used for ICUAW in the main meta-analysis. We assessed publication

bias using a funnel plot.

ICU Rehabilitation and ICU-Acquired Weakness 17/12/2019

Page 7 of 24
RESULTS

Flow of studies through the review

The search yielded 1594 articles. After removing 173 duplicates, we screened 1421 articles for

eligibility (Figure 1). Of these, 41 articles were selected for full-text review. Ten articles[20-

25, 32-35] were selected after the full-text review, but only nine[20-25, 33-35] were included

in the critical appraisal because data could not be obtained for one study,[32] (see reasons for

exclusions: SDC III).

Characteristics of studies

of
Of the 841 patients included in the studies (419 having received an intervention and 422, the

ro
usual care), most received mechanical ventilation during their ICU stay, and the primary reason

for ICU admission and co-morbidities varied from study to study. Details of the patient
-p
characteristics and interventions delivered are shown in the online supplement (SDC II).
re
Effect of intervention

Table 1 shows the number of people with ICUAW in the intervention and control groups as
lP

reported in the articles, in the screened population, and in the total population randomized. The

pooled OR using the random effect model was 0.63 (95% CI: 0.43-0.92) in the screened
na

population, and 0.71 (95% CI: 0.53-0.95) in the total population randomized in favor of early

rehabilitation (Figure 2). Sensitivity analysis using the fixed effect model showed similar results
ur

(plots not shown).

Jo

Subgroup analysis (details shown in Table 2 and Figures 1-4 of the online supplement—SDC

II) showed that the impact of rehabilitation on the odds of developing ICUAW was more

profound in a subgroup of studies in which patients had longer ICU LOS. Sub-analysis, by time

point of ICUAW assessment, favored assessment of ICUAW at hospital discharge compared to

earlier time points. The sub-analysis also showed lower odds of developing ICUAW in studies

ICU Rehabilitation and ICU-Acquired Weakness 17/12/2019

Page 8 of 24
where early rehabilitation was started less than 72 hours compared to studies in which it was

started later than 72 hours. These results were consistent in both the screened population and in

the total randomized population.

The fixed effect model of risk ratio (RR) analysis showed the same results as the OR analysis,

whereas with the random effect model there was only a tendency for a reduced RR in favor of

early rehabilitation: 0.75 (0.57, 1.00) in the screened population and 0.91 (0.82, 1.01) in the

randomized population.

Six studies reported ICU mortality,[20-23, 25, 34] while five studies reported hospital

of
mortality[21, 23, 24, 34, 35]. There was no difference in the pooled OR for acute mortality

ro
between the two groups, 1.19 (95% CI: 0.79-1.80) (results shown in Tables 2 and 3, and Figure

5 of the online supplement—SDC II).

-p
Though ICU LOS was numerically shorter with rehabilitation in six studies,[20, 21, 24, 25, 34,
re
35] only one involving EM was statistically significant (2 days less for ‘LOS until ICU

discharge readiness’)[24] (Table 3). The same study found a significantly shorter hospital LOS
lP

(6.5 days less, p=0.01) in favor of early rehabilitation.[24] Six studies[21, 22, 24, 33-35]

reported median changes in LOS, making it impossible to pool the study results in a meta-
na

analysis.

Length of time on mechanical ventilator was reported as ‘duration of mechanical ventilation’

ur

in eight studies [20-23, 25, 33-35] and as ‘ventilator-free days’ in four studies [22, 24, 25, 35]

(Table 3). Duration of mechanical ventilation was statistically shorter in one study (favoring
Jo

intervention)[35] and had no significant difference in five studies.[20, 22, 23, 25, 34]

Ventilator-free days, was statistically longer in one study (favoring intervention)[35] and

shorter in another study[25] but showed no significant difference in two studies.[22, 24]

Five studies reported discharge location.[23, 24, 33-35] Only two studies compared it

statistically, favoring discharge home in the intervention group p=0.06[35] and 0.0007.[24] A
ICU Rehabilitation and ICU-Acquired Weakness 17/12/2019
Page 9 of 24
meta-analysis of discharge location in the randomized population showed a pooled OR of 1.69

(95% CI: 1.04-2.75) in favor of rehabilitation for being discharged home (SDC II: Table 2 and

Figure 6).

Statistical heterogeneity (I2 statistic) showed values ‘that might not be important’[36] (I2 <

40%) for ICUAW (Fig 2), mortality, and discharge location (SDC II: figure 5- & 6-

supplements, respectively). In contrast, subgroup analysis for ICUAW showed clinical

heterogeneity that may be explained by the type and the timing of rehabilitation interventions

and by ICU LOS. Funnel plot analysis (SDC II: Figure 7-supplement) revealed no significant

of
publication bias, which was also confirmed by Egger's test (P = 0.364 for the null hypothesis

ro
of ‘no small-study effects’).

Details of the risk of bias assessment results are shown in Table 4-supplement (SDC II). With
-p
selection bias, performance bias and detection bias judged as key indicators for our primary
re
outcome,[36] within studies, three studies had high risk of bias, one had an unclear risk, and

five studies showed low risk. Across studies, the risk of bias is summarized as unclear.
lP

DISCUSSION
na

Our study showed that rehabilitation in the ICU is associated with a 37% and 29% reduction in

the odds of developing ICUAW in screened and randomized populations, respectively. This
ur

effect was more pronounced in a subgroup of patients with longer ICU LOS. Our study also

showed that rehabilitation was associated with an increased likelihood of being discharged
Jo

home. There was inconsistent evidence that early rehabilitation is associated with shorter length

of time on mechanical ventilator, and ICU and hospital LOS. It also showed that rehabilitation

is not associated with odds of acute mortality.

ICU Rehabilitation and ICU-Acquired Weakness 17/12/2019

Page 10 of 24
A recent systematic review[19] showed that EM is associated with 58% reduction in the odds

of developing ICUAW. However the robustness of this evidence is not clear as the meta-

analysis in the review pooled only 2 trials excluding the two largest EM trials due on the

review’s exclusion criteria[33] and search date[24]. Two other earlier systematic reviews[17,

18] did not find an association between rehabilitation and reduced incidence of ICUAW which

may be attributed to the limited number of publications[17, 18] (2 and 3 publications,

respectively) included in the reviews. In contrast, this review obtained data from authors of

many primary studies and the meta-analysis pooled nine trials to provide the most roboust

of
evidence to date.

ro
Our sub-analysis showed that a subgroup of studies, in which patients spent a long time in the

ICU, had 49% and 36% lower odds of developing ICUAW with rehabilitation in the screened
-p
and randomized population, respectively. This finding is supported by previous evidence that

showed a greater likelihood of developing ICUAW[33] and greater benefits with exercise in in
re
patients with longer ICU LOS.[37] Our sub-analysis also showed that the effect of rehabilitation
lP

on ICUAW is best assessed at hospital discharge, which is supported by evidence of increased

quadriceps force identified at hospital rather than ICU discharge following a rehabilitation
na

intervention implemented in the ICU.[37] Exercise improves health through neurological,

metabolic and morphological adaptation mechanisms[38-41] which yield measurable effects at

ur

a future time past the point of administration. Assessment at earlier time points may therefore

not accurately reflect the effect of rehabilitation delivered in the ICU. The sub-analysis based
Jo

on the time of onset of rehabilitation suggest that earlier onset of rehabilitation (≤ 72 hours after

ICU admission) is protective against the development of ICUAW when compared to starting

later than 72 hours. However, it may simply imply that patients who are able to participate in

rehabilitation within the first 72 hours of ICU admission, are likely to have less ICUAW. But

the later is less likely as the analysis is based on the timing of rehabilitation across individual

ICU Rehabilitation and ICU-Acquired Weakness 17/12/2019

Page 11 of 24
studies and not for individual patients. Earlier studies have shown decreased skeletal muscle

synthesis and muscle thickness loss within the first 72 hours of critical illness.[42, 43]

Therefore, the timing of the intervention appears to be an important variable to consider.

Our findings also indicated that rehabilitation is safe in the ICU, as has been shown in all the

primary studies included in this review, other primary studies,[37, 44-47] and systematic

reviews.[17, 48-53] Our study showed some evidence of shorter ICU and hospital LOS in one

EM study which is consistent with the results of previous non-RCT studies.[45, 47, 54-56] No

RCT included in this review was powered to detect a difference in ICU or hospital LOS. A

of
previous systematic review[48] pooled the results of several studies together using the Hedges'

ro
g statistic and reported shorter ICU and hospital LOS in favor of rehabilitation; however,

clinical interpretation of the Hedges' g statistic in relation to LOS is unknown. Zhang et. al’s
-p
review[53] showed similar results, but it is unclear how the authors pooled means and medians

together in the meta-analysis. There was also a tendency towards a shorter duration of
re
mechanical ventilator in studies where EM was not part of usual care[25, 35] compared to
lP

studies in which the control group received rehabilitation as part of usual care.[20, 22-24] The

presence of rehabilitation in both the intervention and control groups may be a possible
na

confounder differentiating studies with a positive outcome[25, 35] from those without.[20, 22,

24, 33] In the current review, rehabilitation compared to usual care was also associated with a
ur

higher odds of patients being discharged home. This finding is contrary to the finding of a

previous systematic review which only showed a tendency towards an increased likelihood in
Jo

being discharged home.[52] This difference may be explained by the differences in the studies

included in the review. Contrary to the current review, which included only rehabilitation

studies of EM and NMES that provided information on our primary outcome (ICUAW), the

previous review[52] included both RCTs and controlled study designs that evaluated only

active mobilization interventions.

ICU Rehabilitation and ICU-Acquired Weakness 17/12/2019

Page 12 of 24
The rigorous nature of the search strategy, the inclusion of only RCTs, the conservative nature

of data imputation, the consistency of the results (in the screened and randomized populations,

and with fixed and random effects models for OR) and the ‘absence of evidence of statistical

heterogeneity’ support the strength of the evidence generated in this review. Nonetheless, a

number of limitations exist. First, our search strategy identified only RCTs which provided

information on our primary outcome. Second, sub-analyses may be subject to limitations of

observational investigations, therefore the interpretations of the results of the sub-analysis are

limited. Third, control group interventions varied across studies (with and without EM). We

of
limited the effect of this by focusing on the results of our OR analysis. Finally, ICUAW was

ro
measured with the MRC sum score in all the primary studies which may have underestimated

the real incidence of ICUAW as compared to electrophysiological studies.[57]

-p
In conclusion, our results suggest that beginning rehabilitation early in the course of critical

illness may reduce the odds of developing ICUAW, ICU and hospital LOS, and duration of
re
mechanical ventilation. This is the first robust systematic review to show that rehabilitation in
lP

the ICU reduces the odds of developing ICUAW. Wide-scale adoption of rehabilitation in the

ICU is therefore encouraged. Though such adoption may involve extra cost, potential savings
na

associated with the prevention of ICUAW may imply net savings for the health care system and

improved quality of life for ICU survivors.[45, 56, 58] Studies exploring the return on such
ur

investments are needed. Future studies should also assess ICUAW at hospital discharge, be

adequately powered to detect differences in ICUAW, length of time on mechanical ventilator,

Jo

discharge location, and ICU and hospital LOS.

Ethics approval: Not Applicable

ICU Rehabilitation and ICU-Acquired Weakness 17/12/2019

Page 13 of 24
Source of Funding: This research did not receive any specific grant from funding agencies in

the public, commercial, or not-for-profit sectors.

Conflict of Interest: The authors have no conflict of interest.

Funding: Nil

Systematic review registration number: PROSPERO registration ID: CRD42017065031.

of
Contribution of the Paper

ro
 This systematic review and meta-analysis provides the first evidence that early


-p
rehabilitation in the ICU is associated with lower odds of developing ICUAW.

Our results imply that beginning rehabilitation early in the course of critical illness
re
reduces the odds of developing ICUAW.
lP

Acknowledgements: The authors would like to thank the medical librarian (Pamela Harrison)
na

who provided guidance in the design of the search strategy.

ur

REFERENCES
Jo

1. Griffiths RD, Hall JB. Intensive care unit-acquired weakness. Critical care medicine.
2010;38(3):779-87.
2. Stevens RD, Marshall SA, Cornblath DR, Hoke A, Needham DM, de Jonghe B, et al.
A framework for diagnosing and classifying intensive care unit-acquired weakness. Critical
care medicine. 2009;37(10 Suppl):S299-308.
3. Appleton R, Kinsella J. Intensive care unit-acquired weakness. Continuing Education
in Anaesthesia, Critical Care & Pain. 2012;12(2):62-6.
4. Fan E, Dowdy DW, Colantuoni E, Mendez-Tellez PA, Sevransky JE, Shanholtz C, et
al. Physical complications in acute lung injury survivors: a two-year longitudinal prospective
study. Critical care medicine. 2014;42(4):849-59.

ICU Rehabilitation and ICU-Acquired Weakness 17/12/2019

Page 14 of 24
5. Sidiras G, Gerovasili V, Patsaki I, Karatzanos E, Papadopoulos E, Markaki V, et al.
Short and long term outcomes of ICU acquired weakness. Health science journal. 2013;7(2).
6. Hermans G, Van Mechelen H, Clerckx B, Vanhullebusch T, Mesotten D, Wilmer A,
et al. Acute outcomes and 1-year mortality of intensive care unit-acquired weakness. A cohort
study and propensity-matched analysis. American journal of respiratory and critical care
medicine. 2014;190(4):410-20.
7. Herridge MS, Tansey CM, Matte A, Tomlinson G, Diaz-Granados N, Cooper A, et al.
Functional disability 5 years after acute respiratory distress syndrome. N Engl J Med.
2011;364(14):1293-304.
8. Rudis MI, Guslits BJ, Peterson EL, Hathaway SJ, Angus E, Beis S, et al. Economic
impact of prolonged motor weakness complicating neuromuscular blockade in the intensive
care unit. Critical care medicine. 1996;24(10):1749-56.
9. de Jonghe B, Lacherade JC, Sharshar T, Outin H. Intensive care unit-acquired
weakness: risk factors and prevention. Critical care medicine. 2009;37(10 Suppl):S309-15.
10. Fan E. Critical illness neuromyopathy and the role of physical therapy and

of
rehabilitation in critically ill patients. Respir Care. 2012;57(6):933-44; discussion 44-6.
11. Mehrholz J, Pohl M, Kugler J, Burridge J, Mückel S, Elsner B. Physical rehabilitation
for critical illness myopathy and neuropathy. Cochrane Database of Systematic Reviews

ro
[Internet]. 2015; (3). Available from:
http://onlinelibrary.wiley.com/doi/10.1002/14651858.CD010942.pub2/abstract.
12. Chen YW, Gregory CM, Scarborough MT, Shi R, Walter GA, Vandenborne K.
Transcriptional pathways associated with skeletal muscle disuse atrophy in humans. Physiol
Genomics. 2007;31(3):510-20.
13.
-p
Booth FW. Effect of limb immobilization on skeletal muscle. J Appl Physiol Respir
Environ Exerc Physiol. 1982;52(5):1113-8.
re
14. Wageck B, Nunes GS, Silva FL, Damasceno MCP, de Noronha M. Application and
effects of neuromuscular electrical stimulation in critically ill patients: Systematic review.
Medicina Intensiva. 2014;38(7):444-54.
lP

15. Helliwell TR, Wilkinson A, Griffiths RD, McClelland P, Palmer TEA, Bone JM.
Muscle fibre atrophy in critically ill patients is associated with the loss of myosin filaments
and the presence of lysosomal enzymes and ubiquitin. Neuropath Appl Neuro.
1998;24(6):507-17.
16. Stibler H, Edstrom L, Ahlbeck K, Remahl S, Ansved T. Electrophoretic determination
na

of the myosin/actin ratio in the diagnosis of critical illness myopathy. Intensive Care Med.
2003;29(9):1515-27.
17. Hermans G, De Jonghe B, Bruyninckx F, Van den Berghe G. Interventions for
preventing critical illness polyneuropathy and critical illness myopathy. Cochrane Database of
ur

Systematic Reviews [Internet]. 2014; (1). Available from:

http://onlinelibrary.wiley.com/doi/10.1002/14651858.CD006832.pub3/abstract.
18. Castro-Avila AC, Seron P, Fan E, Gaete M, Mickan S. Effect of Early Rehabilitation
Jo

during Intensive Care Unit Stay on Functional Status: Systematic Review and Meta-Analysis.
PLoS One. 2015;10(7):e0130722.
19. Fuke R, Hifumi T, Kondo Y, Hatakeyama J, Takei T, Yamakawa K, et al. Early
rehabilitation to prevent postintensive care syndrome in patients with critical illness: a
systematic review and meta-analysis. BMJ Open. 2018;8(5):e019998.
20. Dantas CM, Silva PF, Siqueira FH, Pinto RM, Matias S, Maciel C, et al. Influence of
early mobilization on respiratory and peripheral muscle strength in critically ill patients. Rev
Bras Ter Intensiva. 2012;24(2):173-8.
21. Fischer A, Spiegl M, Altmann K, Winkler A, Salamon A, Themessl-Huber M, et al.
Muscle mass, strength and functional outcomes in critically ill patients after cardiothoracic

ICU Rehabilitation and ICU-Acquired Weakness 17/12/2019

Page 15 of 24
surgery: Does neuromuscular electrical stimulation help? The Catastim 2 randomized
controlled trial. Crit Care. 2016;20 (1):1.
22. Kayambu G, Boots R, Paratz J. Early physical rehabilitation in intensive care patients
with sepsis syndromes: a pilot randomised controlled trial. Intensive Care Med.
2015;41(5):865-74.
23. Kho ME, Truong AD, Zanni JM, Ciesla ND, Brower RG, Palmer JB, et al.
Neuromuscular electrical stimulation in mechanically ventilated patients: a randomized,
sham-controlled pilot trial with blinded outcome assessment. J Crit Care. 2015;30(1):32-9.
24. Schaller SJ, Anstey M, Blobner M, Edrich T, Grabitz SD, Gradwohl-Matis I, et al.
Early, goal-directed mobilisation in the surgical intensive care unit: a randomised controlled
trial. Lancet. 2016;388(10052):1377-88.
25. Routsi C, Gerovasili V, Vasileiadis I, Karatzanos E, Pitsolis T, Tripodaki E, et al.
Electrical muscle stimulation prevents critical illness polyneuromyopathy: a randomized
parallel intervention trial. Crit Care. 2010;14(2):R74.
26. Higgins JP, Altman DG, Gotzsche PC, Juni P, Moher D, Oxman AD, et al. The

of
Cochrane Collaboration's tool for assessing risk of bias in randomised trials. BMJ.
2011;343:d5928.
27. Borenstein M, Hedges LV, Higgins JPT, Rothstein HR. A basic introduction to fixed-

ro
effect and random-effects models for meta-analysis. Research Synthesis Methods.
2010;1(2):97-111.
28. Cummings P. The relative merits of risk ratios and odds ratios. Arch Pediatr Adolesc
Med. 2009;163(5):438-45.
29.
-p
van Rhee H, Suurmond R. Meta-analyze dichotomous data: Do the calculations with
Log Odds Ratios and report Risk Ratios or Risk Differences 2015 [Available from: SSRN:
http://dx.doi.org/10.2139/ssrn.3241369.
re
30. van Rhee H, Suurmond R. Meta-analyze dichotomous data: Do the calculations with
Log Odds Ratios and report Risk Ratios or Risk Differences. 2015.
31. Fleiss JL, Berlin JA. Effect sizes for dichotomous data. In: Valentine JC, Hedges LV,
lP

Cooper HM, editors. The handbook of research synthesis and meta-analysis. New York:
Russell Sage Foundation; 2009.
32. Abu-Khaber HA, Abouelela AMZ, Abdelkarim EM. Effect of electrical muscle
stimulation on prevention of ICU acquired muscle weakness and facilitating weaning from
mechanical ventilation. Alexandria Journal of Medicine. 2013;49(4):309-15.
na

33. Denehy L, Skinner EH, Edbrooke L, Haines K, Warrillow S, Hawthorne G, et al.

Exercise rehabilitation for patients with critical illness: a randomized controlled trial with 12
months of follow-up. Crit Care. 2013;17(4):R156.
34. Hodgson CL, Bailey M, Bellomo R, Berney S, Buhr H, Denehy L, et al. A Binational
ur

Multicenter Pilot Feasibility Randomized Controlled Trial of Early Goal-Directed

Mobilization in the ICU. Critical care medicine. 2016;44(6):1145-52.
35. Schweickert WD, Pohlman MC, Pohlman AS, Nigos C, Pawlik AJ, Esbrook CL, et al.
Jo

Early physical and occupational therapy in mechanically ventilated, critically ill patients: a
randomised controlled trial. Lancet. 2009;373(9678):1874-82.
36. Higgins J, Green S. Cochrane handbook for systematic reviews of interventions The
Cochrane Collaboration; 2011. Available from: Available from http://handbook-5-
1.cochrane.org/.
37. Burtin C, Clerckx B, Robbeets C, Ferdinande P, Langer D, Troosters T, et al. Early
exercise in critically ill patients enhances short-term functional recovery. Critical care
medicine. 2009;37(9):2499-505.

ICU Rehabilitation and ICU-Acquired Weakness 17/12/2019

Page 16 of 24
38. Adkins DL, Boychuk J, Remple MS, Kleim JA. Motor training induces experience-
specific patterns of plasticity across motor cortex and spinal cord. J Appl Physiol.
2006;101(6):1776-82.
39. Matta Mello Portugal E, Cevada T, Sobral Monteiro-Junior R, Teixeira Guimarães T,
da Cruz Rubini E, Lattari E, et al. Neuroscience of Exercise: From Neurobiology Mechanisms
to Mental Health. Neuropsychobiology. 2013;68(1):1-14.
40. Rivera-Brown AM, Frontera WR. Principles of exercise physiology: responses to
acute exercise and long-term adaptations to training. PM R. 2012;4(11):797-804.
41. Folland JP, Williams AG. The adaptations to strength training : morphological and
neurological contributions to increased strength. Sports Med. 2007;37(2):145-68.
42. Puthucheary ZA, Rawal J, McPhail M, Connolly B, Ratnayake G, Chan P, et al. Acute
skeletal muscle wasting in critical illness. JAMA. 2013;310(15):1591-600.
43. Schepens T, Verbrugghe W, Dams K, Corthouts B, Parizel PM, Jorens PG. The course
of diaphragm atrophy in ventilated patients assessed with ultrasound: a longitudinal cohort
study. Crit Care. 2015;19(1):422.

of
44. Bailey P, Thomsen GE, Spuhler VJ, Blair R, Jewkes J, Bezdjian L, et al. Early activity
is feasible and safe in respiratory failure patients. Critical care medicine. 2007;35(1):139-45.
45. Morris PE, Goad A, Thompson C, Taylor K, Harry B, Passmore L, et al. Early

ro
intensive care unit mobility therapy in the treatment of acute respiratory failure. Critical care
medicine. 2008;36(8):2238-43.
46. Zafiropoulos B, Alison JA, McCarren B. Physiological responses to the early
mobilisation of the intubated, ventilated abdominal surgery patient. Aust J Physiother.
2004;50(2):95-100.
47.
-p
Muehling B, Schelzig H, Steffen P, Meierhenrich R, Sunder-Plassmann L, Orend KH.
A prospective randomized trial comparing traditional and fast-track patient care in elective
re
open infrarenal aneurysm repair. World J Surg. 2009;33(3):577-85.
48. Kayambu G, Boots R, Paratz J. Physical therapy for the critically ill in the ICU: a
systematic review and meta-analysis. Critical care medicine. 2013;41(6):1543-54.
lP

49. Li Z, Peng X, Zhu B, Zhang Y, Xi X. Active mobilization for mechanically ventilated

patients: a systematic review. Archives of physical medicine and rehabilitation.
2013;94(3):551-61.
50. Silva VS, Pinto JG, Martinez BP, Camelier FWR. Mobilization in the Intensive Care
Unit: systematic review. Fisioterapia e Pesquisa. 2014;21(4):398-404.
na

51. Stiller K. Physiotherapy in intensive care: an updated systematic review. Chest.

2013;144(3):825-47.
52. Tipping CJ, Harrold M, Holland A, Romero L, Nisbet T, Hodgson CL. The effects of
active mobilisation and rehabilitation in ICU on mortality and function: a systematic review.
ur

Intensive care medicine. 2017;43(2):171-83.

53. Zhang G, Zhang K, Cui W, Hong Y, Zhang Z. The effect of early mobilization for
critical ill patients requiring mechanical ventilation: a systematic review and meta-analysis.
Jo

Journal of Emergency and Critical Care Medicine. 2018;2(1).

54. Malkoc M, Karadibak D, Yildirim Y. The effect of physiotherapy on ventilatory
dependency and the length of stay in an intensive care unit. Int J Rehabil Res. 2009;32(1):85-
8.
55. Needham DM, Korupolu R, Zanni JM, Pradhan P, Colantuoni E, Palmer JB, et al.
Early physical medicine and rehabilitation for patients with acute respiratory failure: a quality
improvement project. Archives of physical medicine and rehabilitation. 2010;91(4):536-42.
56. Engel HJ, Tatebe S, Alonzo PB, Mustille RL, Rivera MJ. Physical therapist-
established intensive care unit early mobilization program: quality improvement project for

ICU Rehabilitation and ICU-Acquired Weakness 17/12/2019

Page 17 of 24
critical care at the University of California San Francisco Medical Center. Phys Ther.
2013;93(7):975-85.
57. Appleton RT, Kinsella J, Quasim T. The incidence of intensive care unit-acquired
weakness syndromes: A systematic review. Journal of the Intensive Care Society.
2014;16(2):126-36.
58. McWilliams D, Weblin J, Atkins G, Bion J, Williams J, Elliott C, et al. Enhancing
rehabilitation of mechanically ventilated patients in the intensive care unit: a quality
improvement project. J Crit Care. 2015;30(1):13-8.

of
ro
-p
re
lP
na
ur
Jo

ICU Rehabilitation and ICU-Acquired Weakness 17/12/2019

Page 18 of 24
 of
ro
-p
re
lP
na
ur
Jo

Figure 1. PRISMA Flow Diagram

ICU Rehabilitation and ICU-Acquired Weakness 17/12/2019

Page 19 of 24
 of
ro
-p
Figure 2: Forest Plot of Comparison: Incidence of ICUAW between Early Rehabilitation

versus Usual Care in both the Screened Population and the Randomized Population (timepoint:
re
last ICUAW assessment)
lP
na
ur
Jo

ICU Rehabilitation and ICU-Acquired Weakness 17/12/2019

Page 20 of 24
 f
oo
Table 1: Summary of Results for Primary Outcome- ICUAW

Proportion of events in total

Proportion of events in population randomized
Proportion of events reported screened population (reanalysis*)

pr
Study Time Point Intervention Control p-value Intervention Control Intervention Control
Schweickert 2009 Hospital Discharge 15/49 27/55 0.09 6/40 13/41 Same as Same as
reported reported

e-
Routsi 2010 ICU awakening 3/24 11/28 0.04 3/24 11/28 47/68 55/72

Dantas 2012 Sedation cessation Not reported Not Not 7/14 8/14 19/26 27/33

Pr
reported reported
ICU discharge Not reported Not Not 4/14 7/14 16/26 26/33
reported reported
Denehy 2013 7th-day post-awakening 16/74 13/76 Not 16/57 13/49 33/74 40/76
reported
Kho 2015 ICU awakening 8/12 5/15 0.128 8/12 5/15 12/16 8/18
ICU discharge l 3/12 4/16 1 3/12 4/16 7/16 6/18
na
Hospital discharge 1/12 5/17 0.354 1/12 5/17 5/16 6/18

Kayambu 2015 ICU discharge Not reported Not Not 9/19 14/23 16/26 15/24
reported reported
ur

Fischer 2016 ICU awakening Not reported Not Not

reported reported
ICU discharge Not reported Not Not 6/21 4/20 12/27 11/27
reported reported
Jo

Hospital discharge Not reported Not Not 1/14 1/10 14/27 18/27
reported reported
Hodgson 2016 ICU discharge 7/25 10/20 0.13 7/25 10/20 11/29 11/21
Schaller 2016 ICU Discharge 50/104 51/96 0.95 50/76 51/77 78/104 70/96

ICU Rehabilitation and ICU-Acquired Weakness 17/12/2019

Page 21 of 24
 Table 2. Table of Odds Ratios from Forest Plots

f
(Details of the forest plots are given in supplemental digital content II)

Forest Plot of Comparison:
Incidence of ICUAW between early
rehabilitation versus usual care in
both the screened population and the
oo
Analysis (OR (95% CI)
Screen Population Randomized Population
ICU LOS ≤ 7days for either the ICU LOS not ≤ 7days ICU LOS ≤ 7days for either the ICU LOS not ≤ 7days
intervention or control group for either the intervention or control group for either the
intervention or control intervention or control

pr
randomized population (subanalysis group group
by ICU LOS) 0.96 0.51 0.90 0.64
Figure 1-supplement (0.50-1.85) (0.32-0.81) (0.47-1.72) (0.45-0.90)

e-
Incidence of ICUAW between early ICU 7th day post Sedation ICU Hospital ICU 7th day post Sedation ICU Hospital
rehabilitation versus usual care in Awakening awakening cessation discharge discharge Awakening awakening cessation discharge discharge

Pr
total randomized population 0.92 1.08 0.75 0.78 0.37 1.43 0.72 0.60 0.92 0.54
(subgroup analysis at different (0.05-15.68) (0.46-2.55) (0.17- (0.49- (0.15- (0.28-7.35) (0.38-1.38) (0.17- (0.62- (0.30-
assessment timepoints) 3.33) 1.24) 0.94) 2.08 1.38) 0.97)
Figure 2-supplement

≤ 72 hours ≥72 hours ≤ 72 hours ≥72 hours

Incidence of ICUAW between early
l
na
rehabilitation versus usual care in 0.57 0.70 0.70 0.75
total randomized population (0.37-0.88) (0.17-2.84) (0.50-0.97) (0.42-1.35)
(subgroup analysis by timing of
intervention)
Figure 3-supplement
ur

Incidence of ICUAW between early Early Electrical Both early mobilization & Early Electrical Both early mobilization &
rehabilitation versus usual care in mobilization Stimulation electrical stimulation mobilization Stimulation electrical stimulation
total randomized population 0.71 0.26 0.58 0.70 0.68 0.96
Jo

(subgroup analysis by type of (0.45-1.12) (0.09-0.80) (0.17-1.98) (0.49-1.00) (0.38-1.19) (0.31-3.01)

rehabilitation intervention)
Figure 4-supplement

Mortality between early rehabilitation and usual care Figure 5-supplement 1.19 (0.79-1.80)

Discharge location (to home) between early rehabilitation and usual care Figure 6-supplement 1.69 (1.04-2.75)

ICU Rehabilitation and ICU-Acquired Weakness 17/12/2019

Page 22 of 24
 f
oo
Table 3. Summary of Results for Secondary Outcomes: Length of stay & Length of Time on Mechanical Ventilator

Length of stay Length of Time on Mechanical Ventilator

Mechanical ventilation duration, Ventilator Free Days
Length of stay ICU, days Length of stay hospital, days
days

pr
Usual p- Usual p- Usual Usual p-
Study Rehab. Rehab. Rehab. Rehab.
care value care value care p-value care value
Schweickert 5.9 7.9 0.08 13.5 12.9 0.93 3.4 6.1 0.002 23.5¶ 21.1¶ 0.05
(2.3-7.3) † (4.0-9.6) †

e-
2009 (4.5- (6.1- (8.0-23.1) (8.9-19.8) (7.4-25.6) (0.0-
13.2) † 12.9) † † † † 23.8) †

Routsi 2010 14 22 0.11 Not Not Not 7 10 0.07 4.0 6.0 0.003
(4-62) ‡ (2-92) ‡ (2-41) ꭍ (1-62) ꭍ

Pr
reported reported reported (0.0-16.0) (0.0-
ꭍ 0.41) ꭍ

Dantas 2012 19.41 21.43 0.77 25.12 21.59 0.25 10.24 11.36 0.60 Not Not Not
(10.76) * (17.14) (23.54) * (25.25) * (8.89) * (13.32) * reported reported reported
*
Denehy 2013 8 7l Not 23.54 20.0 Not 4.38s 4.08 Not Not Not Not
na
(6-12) † (6-11) † reported (16.0- (13.0- reported (2.17- (1.98- reported reported reported reported
41.5) † 30.8) † 9.02) †§ 6.69) †§

Kho 2015 22 20 0.72 36 35 0.85 20 16 0.492 Not Not Not

(17) * (17) * (22) * (20) * (18) * (15) * reported reported reported
ur

Kayambu 2015 12 8.5 0.43 41 45 0.80 8 7 0.22 20 21 0.71

(4.0- (3.0- (9-158) † (14-308) (4-64) † (2-30) † (0-24) † (0-26) †
45.0) † 36.0) † †
Jo

Fischer 2016 6 7 0.46 22 19 d = 0.60 2 2 Not Not Not Not

(3-23) ꭍ (3-213) (4-84) ꭍ (9-213) ꭍ (1-7) ꭍ (1-15) ꭍ reported reported reported reported
ꭍ

Hodgson 2016 9 11 0.28 19 (14- 29 0.33 5.4 7 (5.0- 0.18 19.2 (7.4) 17.0 0.4
(6-17) † (8-19) † 30) † (16-34) † (3.5-10) † 12.0) † * (8.7) *
Schaller 2016 7 10 0.0054 15 21.5 0.01 Not Not Not 23.0 22.5 0.31
ICU LOS: (5-12) ꭍ (6-15) ꭍ (11-27) ꭍ reported reported reported

ICU Rehabilitation and ICU-Acquired Weakness 17/12/2019

Page 23 of 24
 (15.0- (18.0- (16.0-
30.0) ꭍ 25.0) ꭍ 25.0) ꭍ

f
oo
*Mean (SD); ‡Mean (range); †Median (interquartile range); ꭍMedian (range); §converted from hrs to days

pr
e-
l Pr
na
ur
Jo

ICU Rehabilitation and ICU-Acquired Weakness 17/12/2019

Page 24 of 24