Professional Documents
Culture Documents
Systematic review
Quebec, Canada
e Epidemiology, Biostatistics and Occupational Health, McGill University, Montreal, Quebec, Canada
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.
Data sources We searched MEDLINE, EMBASE, CINAHL, Cochrane Central and Physiotherapy Evidence Database databases from
inception to May 1st, 2017.
Eligibility criteria Randomized controlled trials of EM and/or NMES interventions in critically ill adults.
Data extraction and data synthesis Data on the incidence of ICUAW and secondary outcomes were extracted. Both odds and risk ratios for
ICUAW were pooled using the random-effects model.
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 trial. Early
rehabilitation decreased the likelihood of developing ICUAW: odds ratio of 0.63 (95% CI: 0.43 to 0.92) in the screened population, and 0.71
(95% CI: 0.53 to 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 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 ventilation duration;
Discharge location
∗ Corresponding author at: CIUSSS du Nord-de-l’Ile-de-Montréal, Hôpital du Sacré-Cœur de Montréal, Axe de recherche en maladies chroniques,
https://doi.org/10.1016/j.physio.2019.12.004
0031-9406/© 2019 Chartered Society of Physiotherapy. Published by Elsevier Ltd. All rights reserved.
2 D.E. Anekwe et al. / Physiotherapy 107 (2020) 1–10
Introduction
Box 1: Inclusion criteria.
Intensive care unit (ICU) survivors often develop muscle Design
weakness which is unrelated to the primary pathology for
• Randomized controlled trials
ICU admission [1]. This weakness may progress to a clini-
cal syndrome known as ICU-Acquired Weakness (ICUAW), Participants
a “clinically detected weakness in critically ill patients in
• Adults
whom there is no plausible etiology other than critical illness”
• Critically ill participants patients not already diag-
[2]. This syndrome includes critical illness polyneuropathy,
nosed with intensive care unit acquired weakness
myopathy, and neuromyopathy [2,3].
(ICUAW)
ICUAW is associated with significant impairments in body
structure and function [4,5], activity limitation [4–6], and par- Intervention
ticipation restriction [4]. It has a long-lasting impact [4–6],
• Evaluated the effect of the early rehabilitation
persisting for months or years following ICU discharge,
interventions of early mobilization (EM) and/or
results in a decline in quality of life for ICU survivors [5,7],
neuromuscular electrical stimulation (NMES) inter-
and places a high burden on the healthcare system [8]. There is
ventions
currently no effective treatment for ICUAW [9,10]. Although
it is often recommended that patients with ICUAW be referred Outcome measures
for rehabilitation, a 2015 Cochrane review [11] failed to
• Reported the incidence of ICUAW or muscle strength
determine whether physical rehabilitation improves activi-
assessed using the Medical Research Council Scale
ties of daily living, muscle strength and quality of life in
(MRC)
these ICU survivors. Thus, preventing the development of
ICUAW through the control of associated risk factors [9] Comparisons
such as immobility [12] is key. Immobility predisposes to
• Usual care
critical illness myopathy through protein loss due to altered
protein metabolism [13] and to critical illness polyneuropa-
thy through hyperglycemia as a result of insulin resistance
[13]. Therefore, reducing the level of immobility during were: to what extent do the early rehabilitation interven-
ICU admission may decrease the risk of developing ICUAW tions of EM and NMES, compared to usual care, reduce the
[9,10]. incidence of ICUAW among patients in the ICU, and alter
Early mobilization (EM) reduces the duration of immo- other outcomes that may be associated ICUAW (i.e. length
bility, while neuromuscular electrical stimulation (NMES) of time on mechanical ventilator, discharge location, ICU and
reduces muscle atrophy [14] which is involved in the patho- hospital length of stay, and acute mortality).
physiological process of ICUAW [15,16]. Both interventions
are sometimes used together as part of an early rehabilitation
intervention in the ICU, however, few robust studies have
evaluated the effectiveness of EM and NMES to reduce the Method
likelihood of developing ICUAW. To date, the few systematic
reviews [17–19] that have evaluated the effectiveness of EM Identification and selection of studies
and NMES to reduce the risk of ICUAW, have included only
two to three RCTs. Additional trials are currently available We searched MEDLINE, EMBASE, CINAHL, Cochrane
[20–24]. Moreover, the meta-analysis in last two systematic Central Register of Controlled Trials and Physiotherapy Evi-
reviews [18,19] analyzed patients with missing data (in some dence Database databases, from inception to May 1st, 2017,
of the included RCTs) as not having ICUAW. In these stud- without language restriction (Supplementary Digital Content
ies, ICUAW was assessed with the MRC sum score, which is [SDC] I). Studies were included if they: (i) were conducted in
prone to having missing data in very weak and uncooperative the ICU, (ii) were RCTs, (iii) involved adult participants, (iv)
patients. Given this context, it seems more appropriate to err evaluated the effect of EM or NMES interventions, and (v)
conservatively on the side of inclusion and impute the worst reported the incidence of ICUAW or assessed muscle strength
values (= ICUAW) for patients who were not evaluable. Fur- using the Medical Research Council Scale (MRC) (Box 1).
thermore, the forest plot in the review by Castro-Avila et al Studies that evaluated rehabilitation interventions in patients
[18] also had errors for the incidence of ICUAW in both the already diagnosed with ICUAW were excluded.
intervention and control groups for the study by Routsi et al. Two reviewers independently screened the titles and
[25]. The few trials included in two previous reviews also lim- abstracts of the retrieved articles, and the full texts of poten-
ited the exploration of sources of heterogeneity. This review tially eligible articles were obtained and further assessed for
is, therefore, necessary to revise and update the knowledge final inclusion. Reviewers formally met at each step to reach a
that is currently available. The specific research questions consensus, and a senior author resolved differences if needed.
D.E. Anekwe et al. / Physiotherapy 107 (2020) 1–10 3
Data analysis
Fig. 2. Forest plot of comparison: incidence of ICUAW between early rehabilitation versus usual care in both the screened population and the randomized
population (timepoint: last ICUAW assessment).
Effect of the intervention 1.80) (results shown in Fig. E of the online supplement—SDC
II).
Table 1 shows the number of people with ICUAW in the Though ICU LOS was numerically shorter with rehabili-
intervention and control groups as reported in the articles, in tation in six studies [20,21,24,25,34,35], only one involving
the screened population, and in the total randomized popula- EM was statistically significant (2 days less for ‘LOS until
tion. The pooled OR using the random effect model was 0.63 ICU discharge readiness’) [24] (Table 3). The same study
(95% CI: 0.43 to 0.92) in the screened population, and 0.71 found a significantly shorter hospital LOS (6.5 days less,
(95% CI: 0.53 to 0.95) in the total population randomized P = 0.01) in favor of early rehabilitation [24]. Six studies
in favor of early rehabilitation (Fig. 2). Sensitivity analysis [21,22,24,33–35] reported median changes in LOS, making
using the fixed effect model showed similar results (plots not it impossible to pool the study results in a meta-analysis.
shown). Length of time on a mechanical ventilator was reported
Details of subgroup analyses are presented in Table 2 and as ‘duration of mechanical ventilation’ in eight studies
Figs. A–D of the online supplement—SDC II. The impact of [20–23,25,33–35] and as ‘ventilator-free days’ in four
rehabilitation on the odds of developing ICUAW was more studies [22,24,25,35] (Table 3). Duration of mechanical
profound in a subgroup of studies in which patients had longer ventilation was statistically shorter in one study (favoring
ICU LOS. Sub-analysis, by time point of ICUAW assessment, intervention) [35] and was not significantly different in five
favored assessment of ICUAW at hospital discharge com- studies [20,22,23,25,34]. Ventilator-free days, was statisti-
pared to earlier time points. The sub-analysis also showed cally longer in one study (favoring intervention) [35], shorter
lower odds of developing ICUAW in studies where early in another study [25] and showed no significant difference in
rehabilitation was started within 72 hours of ICU admission two studies [22,24].
compared to studies in which it was 72 hours after admission. Five studies reported discharge location [23,24,33–35].
These results were consistent in both the screened population Only two studies compared it statistically, favoring discharge
and in the total randomized population. home in the intervention group (P = 0.06 [35] and 0.0007
The fixed effect model of RR analysis showed the same [24]). A meta-analysis of discharge location in the random-
results as the OR analysis, whereas with the random effect ized population showed a pooled OR of 1.69 (95% CI: 1.04
model there was only a tendency for a reduced RR in favor to 2.75) in favor of rehabilitation for being discharged home
of early rehabilitation: 0.75 (0.57, 1.00) in the screened pop- (SDC II: Fig. F).
ulation and 0.91 (0.82, 1.01) in the randomized population. Statistical heterogeneity (I2 statistic) showed values ‘that
Six studies reported ICU mortality [20–23,25,34], while might not be important’ [36] (I2 < 40%) for ICUAW (Fig. 2),
five studies reported hospital mortality [21,23,24,34,35]. mortality, and discharge location (SDC II: Figs. E- & F-online
There was no difference in the pooled OR for acute mor- supplements, respectively). In contrast, subgroup analysis for
tality between the two groups (OR 1.19; 95% CI: 0.79 to ICUAW showed clinical heterogeneity that may be explained
6
Table 2
Table of odds ratios from forest plots (Details of the forest plots are given in the online supplemental digital content II)
A. Forest plot of comparison in the screened population
Table 3
Summary of results for secondary outcomes: length of stay & length of time on mechanical ventilator.
A. ICU and Hospital Length of Stay
Length of stay ICU, days Length of stay hospital, days
Study Rehab. Usual care P-value Rehab. Usual care P-value
Schweickert 2009 [35] 6 (5 to 13)c 8 (6 to 13)c 0.08 14 (8 to 23)c 13 (9 to 20)c 0.93
Routsi 2010 [25] 14 (4 to 62)b 22 (2 to 92)b 0.11 Not Not Not
reported reported reported
Dantas 2012 [20] 19 (11)a 21 (17)a 0.77 25 (24)a 22 (25)a 0.25
Denehy 2013 [33] 8 (6 to 12)c 7 (6 to 11)c Not 24 (16 to 42)c 20 (13 to 31)c Not
reported reported
Kho 2015 [23] 22 (17)a 20 (17)a 0.72 36 (22)a 35 (20)a 0.85
Kayambu 2015 [22] 12 (4 to 45)c 9 (3 to 36)c 0.43 41 (9 to 158)c 45 (14 to 308)c 0.80
Fischer 2016 [21] 6 (3 to 23)d 7 (3 to 213)d 0.46 22 (4 to 84)d 19 (9 to 213)d 0.60
Hodgson 2016 [34] 9 (6 to 17)c 11 (8 to 19)c 0.28 19 (14 to 30)c 29 (16 to 34)c 0.33
Schaller 2016 [24] 7 (5 to 12)d 10 (6 to 15)d 0.005 15 (11 to 27)d 22 (15 to 30)d 0.01
A recent systematic review [19] showed that EM is associ- LOS is unknown. Zhang et al.’s review [53] showed similar
ated with a 58% reduction in the odds of developing ICUAW. results, but it is unclear how the authors pooled means and
However the robustness of this evidence is not clear as the medians together in the meta-analysis. There was also a ten-
meta-analysis in the review pooled only 2 trials excluding dency towards a shorter duration of mechanical ventilation in
the two largest EM trials due on the review’s exclusion cri- studies where EM was not part of usual care [25,35] compared
teria [33] and search date [24]. Two other earlier systematic to studies in which the control group received rehabilitation
reviews [17,18] did not find an association between reha- as part of usual care [20,22–24]. The presence of rehabil-
bilitation and reduced incidence of ICUAW which may be itation in both the intervention and control groups may be
attributed to the limited number of publications [17,18] (2 and a possible confounder differentiating studies with a positive
3 publications, respectively) included in the reviews. In con- outcome [25,35] from those without [20,22,24,33]. In the cur-
trast, this review obtained data from authors of many primary rent review, rehabilitation compared to usual care was also
studies and the meta-analysis pooled nine trials to provide the associated with a higher odds of patients being discharged
most roboust evidence to date. home. This finding is contrary to that reported in a previous
Our sub-analysis showed that a subgroup of studies, in systematic review which only showed a tendency towards
which patients spent a long time in the ICU, had 49% and an increased likelihood of being discharged home [52]. This
36% lower odds of developing ICUAW with rehabilitation in difference may be explained by the differences in the studies
the screened and randomized population, respectively. This included in the review. Contrary to the current review, which
finding is supported by previous evidence that showed a included only rehabilitation studies of EM and NMES rela-
greater likelihood of developing ICUAW [33] and greater tive to our primary outcome (ICUAW), the previous review
benefits with exercise in patients with longer ICU LOS [37]. [52] included both RCTs and controlled study designs that
Our sub-analysis also showed that the effect of rehabili- evaluated only active mobilization interventions.
tation on ICUAW is best assessed at hospital discharge, The rigorous nature of the search strategy, the inclusion
which is supported by evidence of increased quadriceps force of only RCTs, the conservative nature of data imputation, the
identified at hospital rather than ICU discharge following consistency of the results (in the screened and randomized
a rehabilitation intervention implemented in the ICU [37]. populations, and with fixed and random effects models for
Exercise improves health through neurological, metabolic OR) and the ‘absence of evidence of statistical heterogeneity’
and morphological adaptation mechanisms [38–41] which support the strength of the evidence generated in this review.
yield measurable effects at a future time past the point of Nonetheless, a number of limitations exist. First, our search
administration. Assessment at earlier time points may there- strategy identified only RCTs which provided information on
fore not accurately reflect the effect of rehabilitation delivered our primary outcome. Second, sub-analyses may be subject
in the ICU. The sub-analysis based on the time of onset to limitations of observational investigations, therefore the
of rehabilitation suggest that earlier onset of rehabilitation interpretations of the results of the sub-analysis are limited.
(≤72 hours after ICU admission) is protective against the Third, control group interventions varied across studies (with
development of ICUAW when compared to starting later than and without EM). We limited the effect of this by focusing on
72 hours. However, it may simply imply that patients who are the results of our OR analysis. Finally, ICUAW was measured
able to participate in rehabilitation within the first 72 hours with the MRC sum score in all the primary studies which
of ICU admission, are likely to have less ICUAW. But the may have underestimated the real incidence of ICUAW as
later is less likely as the analysis is based on the timing of compared to electrophysiological studies [57].
rehabilitation across individual studies and not for individ- In conclusion, our results suggest that beginning reha-
ual patients. Earlier studies have shown decreased skeletal bilitation early in the course of critical illness may reduce
muscle synthesis and muscle thickness loss within the first the odds of developing ICUAW, ICU and hospital LOS, and
72 hours of critical illness [42,43]. Therefore, the timing of duration of mechanical ventilation. This is the first robust sys-
the intervention appears to be an important variable to con- tematic review to show that rehabilitation in the ICU reduces
sider. the odds of developing ICUAW. Wide-scale adoption of reha-
Our findings also indicated that rehabilitation is safe in the bilitation in the ICU is therefore encouraged. Though such
ICU, as has been shown in all the primary studies included adoption may involve extra cost, potential savings associ-
in this review, other primary studies [37,44–47], and system- ated with the prevention of ICUAW may imply net savings
atic reviews [17,48–53]. Our study showed some evidence for the health care system and improved quality of life for
of shorter ICU and hospital LOS in one EM study which ICU survivors [45,56,58]. Studies exploring the return on
is consistent with the results of previous non-RCT studies such investments are needed. Future studies should consider
[45,47,54–56]. No RCT included in this review was powered evaluating ICUAW at hospital discharge, and should also be
to detect a difference in ICU or hospital LOS. A previous adequately powered to detect differences in ICUAW, length
systematic review [48] pooled the results of several studies of time on mechanical ventilation, discharge location, as well
together using the Hedges’ g statistic and reported shorter as ICU and hospital LOS.
ICU and hospital LOS in favor of rehabilitation; however, Key messages
clinical interpretation of the Hedges’ g statistic in relation to
D.E. Anekwe et al. / Physiotherapy 107 (2020) 1–10 9
• This systematic review and meta-analysis provides the first [12] Chen YW, Gregory CM, Scarborough MT, Shi R, Walter GA, Van-
evidence that early rehabilitation in the ICU is associated denborne K. Transcriptional pathways associated with skeletal muscle
disuse atrophy in humans. Physiol Genomics 2007;31(3):510–20.
with lower odds of developing ICUAW.
[13] Booth FW. Effect of limb immobilization on skeletal muscle. J Appl
• Our results imply that beginning rehabilitation early in the Physiol Respir Environ Exerc Physiol 1982;52(5):1113–8.
course of critical illness reduces the odds of developing [14] Wageck B, Nunes GS, Silva FL, Damasceno MCP, de Noronha M.
ICUAW. Application and effects of neuromuscular electrical stimulation in
critically ill patients: systematic review. Med Intensiva 2014;38(7):
444–54.
Ethics approval: None required. [15] Helliwell TR, Wilkinson A, Griffiths RD, McClelland P, Palmer TEA,
Bone JM. Muscle fibre atrophy in critically ill patients is associated with
Source of funding: This research did not receive any specific the loss of myosin filaments and the presence of lysosomal enzymes
grant from funding agencies in the public, commercial, or and ubiquitin. Neuropath Appl Neuro 1998;24(6):507–17.
not-for-profit sectors. [16] Stibler H, Edstrom L, Ahlbeck K, Remahl S, Ansved T. Electrophoretic
determination of the myosin/actin ratio in the diagnosis of critical illness
Conflict of interest: None declared. myopathy. Intensive Care Med 2003;29(9):1515–27.
[17] Hermans G, De Jonghe B, Bruyninckx F, Van den Berghe G. Interven-
tions for preventing critical illness polyneuropathy and critical illness
myopathy. Cochrane Database Syst Rev 2014;(1). Available from:
Appendix A. Supplementary data http://onlinelibrary.wiley.com/doi/10.1002/14651858.CD006832.
pub3/abstract.
Supplementary material related to this article can be [18] Castro-Avila AC, Seron P, Fan E, Gaete M, Mickan S. Effect of early
rehabilitation during intensive care unit stay on functional status: sys-
found, in the online version, at doi:https://doi.org/10.1016/
tematic review and meta-analysis. PLoS One 2015;10(7):e0130722.
j.physio.2019.12.004. [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.
References [20] Dantas CM, Silva PF, Siqueira FH, Pinto RM, Matias S, Maciel
C, et al. Influence of early mobilization on respiratory and periph-
[1] Griffiths RD, Hall JB. Intensive care unit-acquired weakness. Crit Care eral muscle strength in critically ill patients. Rev Bras Ter Intensiva
Med 2010;38(3):779–87. 2012;24(2):173–8.
[2] Stevens RD, Marshall SA, Cornblath DR, Hoke A, Needham DM, [21] Fischer A, Spiegl M, Altmann K, Winkler A, Salamon A, Themessl-
de Jonghe B, et al. A framework for diagnosing and classifying Huber M, et al. Muscle mass, strength and functional outcomes in
intensive care unit-acquired weakness. Crit Care Med 2009;37(10 critically ill patients after cardiothoracic surgery: does neuromuscular
Suppl):S299–308. electrical stimulation help? The Catastim 2 randomized controlled trial.
[3] Appleton R, Kinsella J. Intensive care unit-acquired weakness. Contin Crit Care 2016;20(1):1.
Educ Anaesth Crit Care Pain 2012;12(2):62–6. [22] Kayambu G, Boots R, Paratz J. Early physical rehabilitation in intensive
[4] Fan E, Dowdy DW, Colantuoni E, Mendez-Tellez PA, Sevransky care patients with sepsis syndromes: a pilot randomised controlled trial.
JE, Shanholtz C, et al. Physical complications in acute lung injury Intensive Care Med 2015;41(5):865–74.
survivors: a two-year longitudinal prospective study. Crit Care Med [23] Kho ME, Truong AD, Zanni JM, Ciesla ND, Brower RG, Palmer
2014;42(4):849–59. JB, et al. Neuromuscular electrical stimulation in mechanically ven-
[5] Sidiras G, Gerovasili V, Patsaki I, Karatzanos E, Papadopoulos E, tilated patients: a randomized, sham-controlled pilot trial with blinded
Markaki V, et al. Short and long term outcomes of ICU acquired outcome assessment. J Crit Care 2015;30(1):32–9.
weakness. Health Sci J 2013;7(2). [24] Schaller SJ, Anstey M, Blobner M, Edrich T, Grabitz SD,
[6] Hermans G, Van Mechelen H, Clerckx B, Vanhullebusch T, Mesotten Gradwohl-Matis I, et al. Early, goal-directed mobilisation in the
D, Wilmer A, et al. Acute outcomes and 1-year mortality of intensive surgical intensive care unit: a randomised controlled trial. Lancet
care unit-acquired weakness. A cohort study and propensity-matched 2016;388(10052):1377–88.
analysis. Am J Respir Crit Care Med 2014;190(4):410–20. [25] Routsi C, Gerovasili V, Vasileiadis I, Karatzanos E, Pitsolis T, Tripo-
[7] Herridge MS, Tansey CM, Matte A, Tomlinson G, Diaz-Granados N, daki E, et al. Electrical muscle stimulation prevents critical illness
Cooper A, et al. Functional disability 5 years after acute respiratory polyneuromyopathy: a randomized parallel intervention trial. Crit Care
distress syndrome. N Engl J Med 2011;364(14):1293–304. 2010;14(2):R74.
[8] Rudis MI, Guslits BJ, Peterson EL, Hathaway SJ, Angus E, Beis S, [26] Higgins JP, Altman DG, Gotzsche PC, Juni P, Moher D, Oxman AD,
et al. Economic impact of prolonged motor weakness complicating et al. The Cochrane Collaboration’s tool for assessing risk of bias in
neuromuscular blockade in the intensive care unit. Crit Care Med randomised trials. BMJ 2011;343:d5928.
1996;24(10):1749–56. [27] Borenstein M, Hedges LV, Higgins JPT, Rothstein HR. A basic intro-
[9] de Jonghe B, Lacherade JC, Sharshar T, Outin H. Intensive care duction to fixed-effect and random-effects models for meta-analysis.
unit-acquired weakness: risk factors and prevention. Crit Care Med Res Synth Methods 2010;1(2):97–111.
2009;37(10 Suppl):S309–15. [28] Cummings P. The relative merits of risk ratios and odds ratios. Arch
[10] Fan E. Critical illness neuromyopathy and the role of physical Pediatr Adolesc Med 2009;163(5):438–45.
therapy and rehabilitation in critically ill patients. Respir Care [29] van Rhee H, Suurmond R, Available from: SSRN: Meta-
2012;57(6):933–44, discussion 44-6. analyze dichotomous data: Do the calculations with Log
[11] Mehrholz J, Pohl M, Kugler J, Burridge J, Mückel S, Elsner Odds Ratios and report Risk Ratios or Risk Differences; 2015
B. Physical rehabilitation for critical illness myopathy and neu- https://doi.org/10.2139/ssrn.3241369.
ropathy. Cochrane Database Syst Rev 2015;(3). Available from:
http://onlinelibrary.wiley.com/doi/10.1002/14651858.CD010942.
pub2/abstract.
10 D.E. Anekwe et al. / Physiotherapy 107 (2020) 1–10
[30] van Rhee H, Suurmond R. Meta-analyze dichotomous data: Do the [44] Bailey P, Thomsen GE, Spuhler VJ, Blair R, Jewkes J, Bezdjian L,
calculations with Log Odds Ratios and report Risk Ratios or Risk et al. Early activity is feasible and safe in respiratory failure patients.
Differences; 2015. Crit Care Med 2007;35(1):139–45.
[31] Fleiss JL, Berlin JA. Effect sizes for dichotomous data. In: Valentine JC, [45] Morris PE, Goad A, Thompson C, Taylor K, Harry B, Passmore L,
Hedges LV, Cooper HM, editors. The handbook of research synthesis et al. Early intensive care unit mobility therapy in the treatment of
and meta-analysis. New York: Russell Sage Foundation; 2009. acute respiratory failure. Crit Care Med 2008;36(8):2238–43.
[32] Abu-Khaber HA, Abouelela AMZ, Abdelkarim EM. Effect of electrical [46] Zafiropoulos B, Alison JA, McCarren B. Physiological responses to
muscle stimulation on prevention of ICU acquired muscle weakness and the early mobilisation of the intubated, ventilated abdominal surgery
facilitating weaning from mechanical ventilation. Alexandria J Med patient. Aust J Physiother 2004;50(2):95–100.
2013;49(4):309–15. [47] Muehling B, Schelzig H, Steffen P, Meierhenrich R, Sunder-Plassmann
[33] Denehy L, Skinner EH, Edbrooke L, Haines K, Warrillow S, Hawthorne L, Orend KH. A prospective randomized trial comparing traditional
G, et al. Exercise rehabilitation for patients with critical illness: a and fast-track patient care in elective open infrarenal aneurysm repair.
randomized controlled trial with 12 months of follow-up. Crit Care World J Surg 2009;33(3):577–85.
2013;17(4):R156. [48] Kayambu G, Boots R, Paratz J. Physical therapy for the critically ill
[34] Hodgson CL, Bailey M, Bellomo R, Berney S, Buhr H, Denehy L, in the ICU: a systematic review and meta-analysis. Crit Care Med
et al. A binational multicenter pilot feasibility randomized controlled 2013;41(6):1543–54.
trial of early goal-directed mobilization in the ICU. Crit Care Med [49] Li Z, Peng X, Zhu B, Zhang Y, Xi X. Active mobilization for mechan-
2016;44(6):1145–52. ically ventilated patients: a systematic review. Arch Phys Med Rehabil
[35] Schweickert WD, Pohlman MC, Pohlman AS, Nigos C, Pawlik AJ, 2013;94(3):551–61.
Esbrook CL, et al. Early physical and occupational therapy in mechan- [50] Silva VS, Pinto JG, Martinez BP, Camelier FWR. Mobiliza-
ically ventilated, critically ill patients: a randomised controlled trial. tion in the intensive care unit: systematic review. Fisioter Pesqui
Lancet 2009;373(9678):1874–82. 2014;21(4):398–404.
[36] Higgins J, Green S, Available from: Cochrane handbook for sys- [51] Stiller K. Physiotherapy in intensive care: an updated systematic review.
tematic reviews of interventions The Cochrane Collaboration; 2011 Chest 2013;144(3):825–47.
http://handbook-5-1.cochrane.org/. [52] Tipping CJ, Harrold M, Holland A, Romero L, Nisbet T, Hodg-
[37] Burtin C, Clerckx B, Robbeets C, Ferdinande P, Langer D, Troosters son CL. The effects of active mobilisation and rehabilitation in ICU
T, et al. Early exercise in critically ill patients enhances short-term on mortality and function: a systematic review. Intensive Care Med
functional recovery. Crit Care Med 2009;37(9):2499–505. 2017;43(2):171–83.
[38] Adkins DL, Boychuk J, Remple MS, Kleim JA. Motor training induces [53] Zhang G, Zhang K, Cui W, Hong Y, Zhang Z. The effect of early
experience-specific patterns of plasticity across motor cortex and spinal mobilization for critical ill patients requiring mechanical ventilation: a
cord. J Appl Physiol 2006;101(6):1776–82. systematic review and meta-analysis. J Emerg Crit Care Med 2018;2(1).
[39] Matta Mello Portugal E, Cevada T, Sobral Monteiro-Junior R, Teix- [54] Malkoc M, Karadibak D, Yildirim Y. The effect of physiotherapy on
eira Guimarães T, da Cruz Rubini E, Lattari E, et al. Neuroscience of ventilatory dependency and the length of stay in an intensive care unit.
exercise: from neurobiology mechanisms to mental health. Neuropsy- Int J Rehabil Res 2009;32(1):85–8.
chobiology 2013;68(1):1–14. [55] Needham DM, Korupolu R, Zanni JM, Pradhan P, Colantuoni E, Palmer
[40] Rivera-Brown AM, Frontera WR. Principles of exercise physiology: JB, et al. Early physical medicine and rehabilitation for patients with
responses to acute exercise and long-term adaptations to training. PM acute respiratory failure: a quality improvement project. Arch Phys
R 2012;4(11):797–804. Med Rehabil 2010;91(4):536–42.
[41] Folland JP, Williams AG. The adaptations to strength training : mor- [56] Engel HJ, Tatebe S, Alonzo PB, Mustille RL, Rivera MJ. Physical
phological and neurological contributions to increased strength. Sports therapist-established intensive care unit early mobilization program:
Med 2007;37(2):145–68. quality improvement project for critical care at the University of Cali-
[42] Puthucheary ZA, Rawal J, McPhail M, Connolly B, Ratnayake G, fornia San Francisco Medical Center. Phys Ther 2013;93(7):975–85.
Chan P, et al. Acute skeletal muscle wasting in critical illness. JAMA [57] Appleton RT, Kinsella J, Quasim T. The incidence of intensive care
2013;310(15):1591–600. unit-acquired weakness syndromes: a systematic review. J Intensive
[43] Schepens T, Verbrugghe W, Dams K, Corthouts B, Parizel PM, Jorens Care Soc 2014;16(2):126–36.
PG. The course of diaphragm atrophy in ventilated patients assessed [58] McWilliams D, Weblin J, Atkins G, Bion J, Williams J, Elliott C,
with ultrasound: a longitudinal cohort study. Crit Care 2015;19(1): et al. Enhancing rehabilitation of mechanically ventilated patients in
422. the intensive care unit: a quality improvement project. J Crit Care
2015;30(1):13–8.
ScienceDirect