Electronic Supplementary Material
ESM Figure E1: Flow diagram of outcome measures search – part one
Abbreviations: Ax, assessment; CINAHL, Cumulative Index to Nursing and Allied Health Literature; EMBASE, the Excerpta Medica Database; ICU, intensive care unit; n, number.
ESM Table E1: Search strategy for part two
Search Strategy for Medline- MH=(“psychometrics”) OR MH=(“observer variation”) OR MH=(“reproducibility of results”) OR MH=(“discriminant analysis”) OR TS=(clin?metr*) OR TS=(valid*) OR TS=(person?metric) OR TS=(reliab*) OR TS=(unidimensional*) OR TS=(discriminability) OR TS=(appropriateness) OR TS=(precision) OR TS=(interpretability) OR TS=(acceptability) OR TS=(practicability) OR TS=(“floor effect”) OR TS=(“ceiling effect”) OR TS=(“standardised response mean”) OR TS=(SRM) OR TS=(“Guyatt’s responsiveness statistic”) OR TS=(“effect size”) OR TS=(“Validation Studies”) OR TS=(“Comparative Study”) OR TS=(“psychometr*”) OR TS=(“clinimetr*”) OR TS=(“observer variation”) OR TS=(reproducib*) OR TS=(“unreliab*”) OR TS=(coefficient) OR TS=(homogeneity) OR TS=(homogeneous) OR TS=(“internal consistency”) OR TS=(“cronbach* alpha*”) OR TS=(“item correlation*”) OR TS=(“item selection*”) OR TS=(“item reduction*”) OR TS=(agreement) OR TS=imprecision OR TS=(“precise values”) OR TS=(“test retest”) OR TS=(“test-retest”) OR TS=(“reliab* test”) OR TS=(“reliab* retest”) OR TS=stability OR TS=interrater OR TS=inter-rater OR TS=intrarater OR TS=intra-rater OR TS=intertester OR TS=inter-tester OR TS=intratester OR TS=intra-tester OR TS=interobserver OR TS=interobserver OR TS=intraobserver OR TS=intraobserver OR TS=intertechnician OR TS=inter-technician OR TS=intratechnician OR TS=intra-technician OR TS=interexaminer OR TS=inter-examiner OR TS=intraexaminer OR TS=intra-examiner OR TS=interassay OR TS=inter-assay OR TS=intraassay OR TS=intra-assay OR TS=interindividual OR TS=inter-individual OR TS=intraindividual OR TS=intra-individual OR TS=interparticipant OR TS=inter-participant OR TS=intraparticipant OR TS=intra-participant OR TS=kappa OR TS=kappa’s OR TS=kappas OR TS=(repeatab*) OR TS=(“replicab* measure*”) OR TS=(“replicab* findings”) OR TS=(“replicab* result*”) OR TS=(“replicab* test*”) OR TS=(“repeated measure*”) OR TS=(“repeated findings”) OR TS=(“repeated result*”) OR TS=(“repeated test*”) OR TS=generaliza* OR TS=generalisa* OR TS=concordance OR TS=(“intraclass correlation*”) OR TS=discriminative OR TS=(“known group”) OR TS=(“factor analysis”) OR TS=(“factor analyses”) OR TS=dimension* OR TS=subscale* OR TS=(“multitrait scaling analys*”) OR TS=(“item discriminant”) OR TS=(“interscale correlation*”) OR TS=error* OR TS=(“individual variability”) OR TS=(“variability analys*”) OR TS=(“variability values”) OR TS=(“uncertainty measurement) OR TS=(“uncertainty measuring) OR TS=(“standard error of measurement”) OR TS=(sensitiv*) OR TS=responsive* OR TS=("minimal* detectable change") OR TS=(MDC) OR TS=(“minimal* clinical* important* difference”) OR TS=(“minimal* clinical* significant difference”) OR TS=(“minimal* clinical* detectable* difference”) OR TS=(“minimal* clinical* important* change”) OR TS=(“minimal* clinical* significant change”) OR TS=(“minimal* clinical* detectable* change”) OR TS=(“clinical* important* difference”) OR TS=(“clinical* significant difference”) OR TS=(“clinical* detectable* difference”) OR TS=(“clinical*important* change”) OR TS=(“clinical* significant change”) OR TS=(“clinical* detectable* change”) OR TS=(“small detectable change”) OR TS=(“small real change”) OR TS=(“small detectable difference”) OR TS=(“small real difference”) OR TS=(“clinical* important*difference”) OR TS=(“clinical* significant difference”) OR TS=(“clinical* detectable* difference”) OR TS=(“clinical*important* change”) OR TS=(“meaningful change”) OR TS=(“Item response model”) OR TS=IRT OR TS=Rasch OR TS=(“Differential item functioning”) OR TS=DIF OR TS=(“Computer adaptive testing”) OR TS=(“item bank”) OR TS=(“cross-cultural equivalence”)
- TS="intensive care unit*" OR TS="critical care" OR TS=ICU OR TS="crit* ill*" OR TS=sepsis OR TS=MODS OR TS="multi* organ dysfunction syndrome*"
- Each group run separately as described below:
- TS=“skinfold thickness*” OR TS=“mid arm circumference*” OR TS=“mid-arm circumference*” OR TS=“MAC” OR TS=“anthropometry*” OR TS=“BIS” OR TS=“body composition*” OR TS=“bioimpedance spectroscopy*” OR TS=“electric* impedance*” OR TS=“BIA”
- TS=“ultrasound*” OR TS=“sonography” OR TS=“US” OR TS=“diagnostic ultrasound*” AND (TS=muscle OR TS=quad*)
- TS=“circumference*” NOT TS=“waist circumference”
- 1 OR 2 OR 3
TS="manual muscle test*" OR TS=MMT OR TS="manual strength test*" OR TS= “Kendall muscle test*” OR TS= “Kendall scale*” OR TS= “muscle strength grade*” OR TS="Oxford muscle" OR TS="Oxford scale" OR TS="oxford muscle grade" OR TS="muscle strength* grade scale"OR TS= “muscle strength” OR TS=MRC OR TS= “MRC sum-score” OR TS= “MRC sum score*” OR TS= “Medical research council” OR TS= “Worthingham’s muscle test*” OR TS= “hand-held dynamomet*” OR TS= “handheld dynamomet*” OR TS= “hand held dynamomet*” OR TS= “hand dynamomet*” OR TS="hand-grip dynamomet*" OR TS="handgrip dynamomet*" OR TS="hand grip dynamomet*" OR TS="hand strength dynamomet*" OR TS="Jamar dynamomet*" OR TS="grip dynamomet*" OR TS="hand dynamomet*" OR TS=hand-held dynamomet* OR TS=handheld dynamomet* OR TS=hand held dynamomet* OR TS=HHD OR TS=dynamomet*OR TS=dynamomet* OR TS=“hand strength*” OR TS= “Jamar dynamomet*” OR TS=“sit-to-stand*” OR TS=sit to stand*” OR TS=“chair-stand test*” OR TS=“chair stand test*” OR (TS=grip AND TS=strength)
Function
TS=“exercise capacity” OR TS=“functional capacity” OR TS=6MWT OR TS=6MWD OR TS=6MW OR TS=“six minute walk* test*” OR TS=“six-minute walk* test*” OR TS=“6-minute walk* test*” OR TS=“6 minute walk* test*” OR TS=“six min*”OR TS=“six-min*” OR TS=“6 min*” OR TS=6-min* OR TS=6-MWT OR TS=6-MWD OR TS=6-MW OR TS=“6 MWT” OR TS=“6 MWD” OR TS=“6 MW” OR TS=“ISWT” OR TS=“shuttle-walk*-test*” OR TS=“shuttle-walk test*” OR TS=“shuttle walk* test*” OR TS=“incremental-shuttle walk*” OR TS=“shuttle walk*” OR TS=“shuttle walk*” OR TS=“TUG” OR TS=“timed-up-and-go test” OR TS=“timed up and go test*” OR TS=Ten-metr* OR TS=“10 metr*” OR TS=“10-metr*” OR TS=ten-metre-walk* OR TS=“10 metre-walk*” OR TS=“10-metr-walk*” OR TS=ten-meter-walk* OR TS=“10 meter-walk*” OR TS=“10-meter-walk*” OR TS=four-metr* OR TS=“4 metr*” OR TS=“4-metr*” OR TS=four-metre-walk* OR TS=“4 metre-walk*” OR TS=“4-metr-walk*” OR TS=four-meter-walk* OR TS=“4 meter-walk*” OR TS=“4-meter-walk*” OR TS=metre-walk* OR TS=“metre walk*” OR TS=“metr-walk*” OR TS=meter-walk* OR TS=“walk* test*” OR TS=“walk-test*” OR TS=“Barthel Index” OR TS=“Barthel” OR TS=“FIM” OR TS=“Functional independence measure*” OR TS=“mm-FIM” OR TS=“mini-modified FIM” OR TS=“modified FIM” OR TS=“FSS-ICU” OR TS=“Functional status score*” OR TS=“RMI” OR TS=“Rivermead mobility index” OR TS=“SOMS scale” OR TS=“surgical optimal mobility scale*” OR TS=“functional milestone*” OR TS=“ICU Mobility scale*” OR TS=“PFIT” OR TS=“physical function in intensive care test*” OR TS=“modified Rankin” OR TS=“Rankin scale” OR TS=“Chelsea Critical Care Physical Assessment tool” OR TS=“CPAx” OR TS=“Lawton index” OR TS=“IADL” OR TS=“Katz” OR TS=“ADL” OR TS=“Katz ADL” OR TS=“modified Katz” OR TS=“Modified ADL” OR TS=“2MWT” OR TS=“2MWD” OR TS=“2MW” OR TS=“two minute walk* test*” OR TS=“two-minute walk* test*” OR TS="2-minute walk* test*" OR TS="2 minute walk* test*" OR TS=“2-MWT” OR TS=“2-MWD” OR TS=“2-MW” OR TS="2 MWT" OR TS="2 MWD" OR TS="2 MW" OR TS=“Perme mobility scale*” OR TS=“University of Rochester Acute Care Evaluation” OR TS=“URACE” OR TS=“Berg Balance scale” OR TS=“BBS” OR TS=“Balance scale” OR TS=“FAC” OR TS=“Functional ambulatory categ*” OR TS=“Fried’s 5-point scale” OR TS=“Fried’s frailty index” OR TS=“frailty” OR TS=“Kansas University Hospital Physical Therapy Acute Care Functional Outcomes Tool” OR TS=“Functional disability scale” OR TS=“FDS”OR “Elderly mobility scale”
ESM Table E2: Methodological quality of included studies in part two
Author, year / Reliability / Measurement error / Hypothesis testing / Criterion validity / ResponsivenessMuscle mass
Baldwin 2014[1] / Poor / X / Poor / X / X
Grimm 2013[2] / Poor / X / Poor / X / Poor
Savalle 2012[3] / X / Fair / Fair / X / X
Baldwin 2012[4] / Poor / X / X / X / X
Gruther 2008[5] / X / X / Fair / X / X
Reid 2004[6] / X / X / Poor / X / Poor
Faisy 2000[7] / X / X / Poor / Poor / X
Frankenfield 1999[8] / X / X / Poor / X / X
Morais 1998[9] / X / X / X / Poor / X
Phang 1996[10] / X / X / Fair / X / Fair
Campbell 1995[11] / X / X / Poor / X / X
Robert 1993[12] / X / X / Poor / X / X
Muscle strength
Yosef Brauner 2014[13] / X / X / Poor / X / X
Fan 2013[14] / X / X / Fair / X / x
Connolly 2013[15] / Poor / X / Fair / Good / X
Baldwin 2013 [16] / Poor / Poor / X / X / Poor
Lee 2012 [17] / X / X / Good / Good / X
Hermans 2012 [18] / Good / X / X / X / X
Vanpee 2011[19] / Good / X / X / X / X
Hough 2011 [20] / Fair / X / X / X / X
Fan 2010 [21] / Poor / X / X / X / X
Brunello 2010[22] / X / X / Poor / Fair / X
Sharshar 2009[23] / X / X / X / Excellent / X
Ali 2008 [24] / Poor / X / Good / Excellent / X
Functional outcomes
Denehy 2014 [25] / X / X / Good / X / X
Corner 2014 [26] / X / X / Good / X / X
Tripathy 2014 [27] / X / X / X / Good / X
Baldwin 2014, frailty [28] / X / X / Poor / Poor / X
Nawa 2014 [29] / Poor / X / X / X / X
Nordon-Craft 2014[30] / X / X / Good / Good / Good
Hodgson 2014 [31] / Good / X / X / X / X
Denehy 2013[32] / X / X / Fair / Excellent / Fair
Corner 2013 [33] / Poor / X / Fair / X / X
Abd-El-Gawad 2013[34] / X / X / X / Good / X
Thrush 2012[35] / X / X / Poor / X / Poor
Kasotakis 2012 [36] / Fair / X / Good / Good / X
Allison 2012[37] / X / Good / Fair / X / Good
Vest 2011[38] / X / X / Fair / X / X
Daubin 2011[39] / X / X / Fair / X / X
Clini 2011[40] / X / X / Fair / Fair / Fair
Skinner 2009[41] / Poor / Poor / X / X / Poor
Sacanella 2009[42] / X / X / X / Excellent / X
Van der Schaaf 2008[43] / X / X / Poor / Fair / X
Swafford 2008[44] / X / X / Poor / X / X
Chiang 2006[45] / Poor / X / Fair / X / Fair
Bo 2003[46] / X / X / X / Excellent / X
Dardaine 2001 [47] / X / X / X / Excellent / X
Abbreviations: x; not assessed.
ESM Table E3: Results - Inter-rater-rater reliability, intra-rater reliability and measurement error of outcome measures
Author, year / Type of reliability and outcome measure / Time interval / Reliability coefficient / Measurement errorMUSCLE MASS
Baldwin 2014[1] / Intra-rater reliability 1. US (thickness) and 2. Strength (HGD, HHD) / Tests repeated within one day of first test / Muscle thickness: overall ICC0.98
Muscle strength: overall ICC0.83
Grimm 2013[2] / 1. Inter-rater
2. Intra-rater for Heckmatt US echogenicity / NR / 1. ICC=0.92
2. ICC=0.97
Savalle 2012[3] / 1. Measurement error BIS for resistance (R), reactance (Xc) at two-timepoints / R1 =2.1
R2=4.8
Xc1=4.2
Xc2=2.9
Baldwin 2012[4] / 1. Intra-session reliability and
2. Test-retest reliability for BIS / Tests repeated by same investigator 2 days later, 30 secs between 3 BIS measures / 1. SFB7 (Tetrapolar device) ICC [95%CI] range from 0.99 [0.99-1.00] to 1.00 [1.00-1.00], p≤0.001 and Inbody S20 (Octopolar device) ICC [95%CI] range from 0.44 [0.14-0.73] to 0.96 [0.90-0.98]
2. SFB7 (Tetrapolar device) ICC range [95%CI]=0.94 [0.83-0.98] to 0.99 [0.97-1.0], p≤0.001 and Inbody S20 (Octopolar device) ICC [95%CI] range=-0.21 [-2.04-0.56] to 0.96 [0.88-0.99]
MUSCLE STRENGTH
Connolly 2013[15] / 1. Inter-rater reliability and agreement for Dx of ICU-AW for (a) simulated and (b) ICU patients
2. Inter-rater reliability for individual muscle group scores for (a) simulated and (b) ICU patients / 30 minutes between tests / 1 (a) Simulated patients: ICC [95%CI]=1.0 [0.99-1.0], Kappa [95%CI]=1.0 [1.0-1.0] and (b) ICU patients: ICC [95%CI]=0.94 [0.85-0.98], Kappa [95%CI]=0.60 [0.25-0.95]
2 (a) ICC range from 0.93 to 1.0 and (b) ICC range from 0.63 to 0.88
Hermans 2012 [18] / Inter-rater reliability for
1. MRC,
2. Dx of ICU-AW (<48/60, < 36/60); and 3. HGD / Between 30 mins to 2-hrs between tests / 1. ICC [IQR]=0.95 [0.92-0.97], weighted Kappa mean (SE)=0.83± 0.03
2. MRC < 48 Kappa (SE)=0.68 (0.09); MRC < 36 Kappa mean (SE)=0.93 (0.07)
3. Right HGD ICC [IQR}=0.93 [0.86-0.97] and left HGD ICC [IQR]=0.97 [0.94-0.98]
Baldwin 2013[16] / Inter-rater, test-retest reliability and SEM for right sided 1. HGD; 2. HHD- quads; 3. HHD-biceps / 2-4-hrs between tests for inter-rater reliability.
Test-retest repeated two days post first test / Inter-rater reliability ICC (agreement) [95%CI] reported for following:
1. 0.92 [0.68-0.98]
2. 0.84 [0.52-0.95]
3. 0.71 [-0.21-0.93]
Test retest reliability ICC (agreement) [95%CI] reported for following:
1. 0.92 [0.78-0.97]; 2. 0.90 [0.71-0.96]; 3. 0.82 [0.39-0.94] / 1. SEM=2.8kg
2. SEM=2.6kg
3. SEM=1.9kg
Vanpee 2012[19] / Inter-rater reliability for HHD –multiple muscle groups / At least 2 hrs between tests (on same day) / ICC [95%CI] reported for the following:
Shd Abd=0.91 [0.85-0.95]; Elb flex=0.96 [0.93-0.98]
Wrist Ext=0.94 [0.91-0.97]; Hip Flex=0.80 [0.67-0.89]
Knee Ext=0.94 [0.90-0.97]; Ankle DF=0.76 [0.33-0.90]
Hough 2011 [20] / Inter-rater agreement for
1. MRC,
2. Individual muscle groups
2. Diagnosis of ICU-AW / 30 minutes between tests / 1. ICC [95%CI]=0.83 [0.67-0.93]
2. ICC range = 0.31-0.75
3. Kappa=0.38 in ICU, Kappa=1.0 in hospital ward
Fan 2010[48] / Inter-rater reliability for 1. MRC-SS; 2. Composite UL and LL scores and 3. Agreement for Dx of ICU-AW for two groups (a) simulated and (b) ICU cohort / NR / 1. (a) ICC range=(0.69-1.0); (b) ICC range=(0.86-1.0)
2. (a) ICC range=(0.62-1.0) in UL and [0.66-1.0) in LL respectively; (b) ICC range=(0.45-1.0) for UL and (0.98-1.0) in LL respectively
3. (b only): all muscle groups Kappa [95%CI]=0.88 [0.44-1.0]; UL only Kappa [95%CI]=0.88 [0.44-1.0] and LL only Kappa [95%CI]=1.00 [0.55-1.0].
Ali 2008 [24] / Inter-rater and test-retest reliability for Dx of ICU-AW (n=12) / < 4-hrs between tests / 1. Inter-rater reliability: complete agreement and average MRC scores correlation r=0.96, p<0.001
2. Test-retest reliability: agreement for classification of ICU-AW yes/no Kendall’s tau-b=0.96, p<0.001 and average MRC scores correlated rho=0.90, p<0.001
FUNCTION
Nawa 2014[29] / Inter-rater reliability for 1. Overall score and 2. Individual items on Perme mobility scale / Simultaneous testing / 1. Overall score ICC [95%CI]=0.99 [0.98-1.0]
2. Individual items kappa range=0.60-1.00
Hodgson 2014 [31] / Inter-rater for ICU mobility scale / Testing performed within 30 mins of each assessor / 1. Overall ICC [95%CI]=0.80 [0.75-0.84]
Corner 2014[26] / Inter-rater for scoring of CPAx 1. Overall score and 2. Individual items (n=3) / Simultaneous testing / 1. kappa [95%CI]=0.99 [0.79-1.0], p<0.001 and Cronbach’s alpha=0.80
2. ICC range=[0.97-1.00]
Kasotakis 2014 [36] / Inter-rater agreement for SOMS scoring / Simultaneous testing / 1. Kappa [95%CI]=0.80 [0.70-0.90]
Skinner 2009 [41] / Inter-rater for scoring of PFIT items 1. MOTS cadence and 2. bilateral shoulder flexion cadence / Simultaneously testing / 1. ICC [95%CI]=1.0 [0.99-1.0]
2. ICC [95%CI]=1.0 [0.97-1.0] / 1. MOTS cadence SEM = 0.76
2. Bilateral shd flex cadence (reps/min) SEM = 0.61
Chiang 2006[45] / 1. Inter-rater and 2. Intra-rater reliability HHD (n=5) / NR / 1. ICC=0.83
2. ICC=0.83
Abbreviations: Abd, abduction; BIS, bioimpedance spectroscopy; CPAx, Chelsea Physical Assessment Tool; DF, dorsiflexion; Dx, diagnosis; Elb, elbow; ext; extension; Flex, flexion; HGD, handgrip dynamometry; HHD, handheld dynamometry; hrs, hours; ICC, intraclass correlation coefficient; ICU, intensive care unit; ICU-AW, intensive care unit acquired weakness; LL, lower limb; LOA, limits of agreement; m, metres; mins; minutes; MOTS, marching on the spot; MRC, medical research council; MRC-SS, medical research council sum-score; NR, not reported; PFIT, Physical function in intensive care test; R, resistance; reps, repetitions; SE, standard error; SEM, standard error measurement; secs; seconds; shd, shoulder; SOMS, surgical optimal mobility scale; UL, upper limb; US, ultrasound; Xc, reactance; 6MWD, six minute walk distance; 6MWT, six minute walk test; 95%CI, 95 percent confidence intervals.
ESM Table E4: Results - Criterion-concurrent validity, criterion-predictive validity and construct validity of outcome measures
Author, year / Type of validity and outcome measure / Comparator outcome measure or predicted outcome / Validation resultsMUSCLE MASS
Baldwin 2014[1] / Construct validity US (muscle thickness) and BIS (FFM) / 1. Biceps thickness vs biceps strength
2. Forearm thickness vs HGD
3. Quads thickness vs Quads strength
4. FFM (BIS) vs biceps, forearm and quads thickness respectively / 1. r2=0.43, p=0.006; 2. r2=0.52, p=0.002; 3. r2=0.22, p=0.07; 4. r2=0.30; p=0.03; r2=0.37, p=0.02; r2=0.61, p≤0.001
Grimm 2013 [2] / Construct validity US echogenicity / 1. US fasciculation score at D4 and D14 / 1. D4: (r=0.4, p=0.001) and D14 (r=0.62, p<0.001)
Savalle 2012[3] / Construct validity circumference / 1. Leg and arm circumference / 1. Spearman rho=0.59
Gruther 2008[5] / Construct validity (a) US (thickness) (b) circumference / 1. LOS / 1a. US study 1: baseline (L) and (R) thigh thickness negatively correlated with ICU LOS (p=0.004) and (p=0.005) respectively and diff from D1 to D28 (R) (p=0.01) and (L) (p=0.003); Study 2: Baseline (L) and (R) thigh thickness negatively correlated with ICU LOS (p<0.001)
1b. Thigh circumference study 2: both legs negatively correlated with LOS (p=0.003)
Reid 2004 [6] / Construct validity US (thickness) / 1. Daily energy balance / 1. r=-0.14, p=0.51
Faisy 2000[7] / Construct and criterion predictive validity BIS and criterion predictive MAC only / 1. Plasma albumin levels
2. Prealbumin levels
3. ICU mortality / 1. BIS (ACM/BW ratio) and plasma albumin levels r=0.38, p=0.005; BIS (ECW/ICW) ratio and plasma albumin levels r=0.43, p=0.001
2. BIS (ACM/BW ratio) and plasma prealbumin levels r=0.32, p=0.01; BIS (ECW/ICW) ratio and plasma prealbumin levels r=0.27, p=0.04
3. Individuals with malnutrition as defined by BIS (ACM/BW ratio ≤40.6% weight) have higher mortality (p=0.01) and BIS (ECW/ICW ratio) NS with mortality (p>0.05); individuals with normal MAC have higher ICU mortality compared to those with low MAC (p=0.01);
Frankenfield 1999 [8] / Construct validity BIS / 1. Resting metabolic rate
2. O2 consumption / 1. BIS (FFM) to resting metabolic rate r2=0.47 p<0.0001
2. BIS (BCM) to O2 consumption r2=0.75, p<0.001
Morais 1998[9] / Criterion predictive validity BIS / 1. In-hospital mortality / 1. NS difference between survivors and deceased for resistance, reactance, body impedance co-efficient (p>0.05)
Phang 1996[10] / Construct validity BIS (ECM) / 1. Change in ECM from baseline to D7 compared to 7-day energy balance
2. Change in ECM from baseline to D7 compared to 7 day fluid balance / 1. No correlation with 7-day energy balance r<0.1, p<0.5
2. r=0.38, p<0.01
Campbell 1995[11] / Construct validity US (muscle thickness) and MAC / 1. FFM derived from skinfold thickness measurements / 1. Biceps thickness r=0.84; triceps r=0.56; forearm r=0.73; thigh r=0.66; calf r=0.66; rectus r=0.64
MAC r=0.74
Robert 1993[12] / Construct validity BIS (BCM) / 1. Change in BCM compared to energy intake
2. Change in BCM and protein intake / 1. r2=0.87, p<0.001
2. r2=0.65, p<0.001
MUSCLE STRENGTH
Yosef-Brauner 2014 [13] / Construct validity MRC / 1. Baseline sitting Balance
2. Baseline right HGD
3. Average change in MRC in relation to average change in MIP / 1. r=0.67
2. r=0.62
3. r=-0.62
Fan 2013[14] / Construct validity HGD, MRC, MAC / 1. HGD vs MIP
2. MRC vs MIP and HGD
3. MAC vs HGD and MRC / 1. r2=0.89
2. r2=0.94 and r2=0.99
3. Decrease in HGD at all time points r2=0.97 and decrease in MRC at all time points r2=NR
Connolly 2013 [15] / Criterion predictive and construct validity MRC / 1. ICU and hospital mortality
2. ICU and hospital LOS
3. HGD at ICU DC
4. Barthel at ICU DC
5. EMS at ICU DC / 1. No association with ICU (p=0.67) or hospital mortality (p=0.53)
2. Significant association with ICU (p=0.004) and hospital LOS (p=0.04). MRC<48/60 at awakening had high sensitivity for ICU (93%) and hospital LOS (84%); poor specificity for ICU (41%) and hospital LOS (41%) and poor PPV for ICU (54%) and hospital LOS (67%)
3. (L) HGD r=0.5, p=0.0003 and (R) HGD r=0.5, p < 0.0001
4. r=0.4, p=0.005; 5. r=0.4, p=0.005
Lee 2012 [17] / Criterion predictive and construct validity MRC and HGD / 1. In-hospital mortality
2. ICU LOS
3. Hospital LOS
4. MV duration
5. MRC and HGD
6. HGD and above (1 to 4) / Univariate analyses
1. MRC and mortality: OR=0.94, p=0.006; HGD and mortality: OR=1.0, p=0.74; predictive ability of MRC for mortality AUC=0.73.
2. MRC and ICU LOS:
r=-0.31, p=0.002; HGD and ICU LOS: r=0.06, p=0.55
3. MRC and hospital LOS:
R=-0.33, p=0.001; HGD and hospital LOS: r=-0.05; p=0.65
Multivariate analyses:
1. MRC independently associated with mortality OR=0.95, p=0.04
2. MRC independently associated with ICU LOS IRR=0.98, p< 0.001
3. MRC and hospital LOS IRR=0.97, p<0.001
4. MRC and MV duration p=0.01
5. HGD and MRC scores correlated r=0.55, p<0.0001
6. HGD not independently associated with outcomes p>0.05
Brunello 2010 [22] / Criterion predictive and construct validity MRC (Dx of ICU-AW) / 1. 180-day survival
2. Barthel Index
3. TISS-28
4. MV days
5. ICU LOS / 1. Survival significantly higher in patients with no ICU-AW 77% vs ICU-AW 38%, p=0.03
2. Barthel Index lower in patients with ICU-AW mean (SD)=43±21 vs no ICU-AW mean (SD)=81±24, p=0.01
3. TISS-28 scores at admission, D3, and D7 not significantly different between patients with ICU-AW and no ICU-AW (p>0.05)
4. ICU-AW ventilated longer mean±SD=14±5 vs no ICU-AW mean±SD=7±5, p=0.001
5. ICU-AW cohort had longer ICU LOS mean±SD=17±6 vs no ICU-AW mean±SD=10±6, p=0.005
Sharshar 2009[23] / Criterion predictive validity awakening MRC score and Dx of ICU-AW / 1. ICU mortality
2. In-hospital mortality / After adjusted outcome analysis
1. For each MRC point decrease ICU mortality increased by OR[95%CI]=1.05 [1.01-1.09], p=0.006
Dx of ICU-AW OR[95%CI]=7.99 [0.99-64.29], p=0.05
2. For each MRC point decrease in-hospital mortality increased by OR[95%CI]=1.03 [1.01-1.05] , p = 0.03
Dx of ICU-AW OR[95%CI]=2.02 [1.03-8.03], p=0.05
Swafford 2008[44] / Construct validity change in total KUH score / 1. Change in transfer and gait FIM / 1. r=0.86
Ali 2008 [24] / Criterion predictive and construct validity MRC and HGD (both for Dx of ICU-AW) / 1. Mortality
2. ICU and hospital LOS
3. MRC and HGD / 1. MRC: OR[95%CI]=7.80 [2.40-25.30] p=0.001 after adjusting for severity of illness and organ failures
HGD: OR[95%CI]=4.50 [1.50-13.60], p=0.007
2. MRC: Relative reduction in ICU free days (p=0.001) and hospital free days (p=0.007)
HGD: Relative reduction in ICU free days (p=0.001) and hospital free days (p=0.07)
3. HGD (for Dx of ICUAW) had good test performance (overall sensitivity 81%, specificity 83%, NPV 92% and PPV 63%) when compared with an ICU-AW Dx by MRC. Performance NS different between genders
Correlation between maximum HGD and average MRC score for each muscle group rho=0.64, p<0.001
FUNCTION
Corner 2014[26] / Construct validity ICU discharge CPAX / 1. Hospital discharge destination / 1. Significant difference between groups for discharge destination (p<0.0001)
Denehy 2014 [25] / Construct validity ICU discharge 6MWT and TUG test / 1. Berg Balance Scale
2. 5x STS test
3. SF-36 PF domain
4. TUG test (only for 6MWT) / 6MWT
1. rho =0.80 p<0.001
2. rho=-0.69, p<0.001
3. rho=0.69, p<0.001
4. rho=-0.79, p<0.001
TUG test
1. rho=-0.91, p<0.001
2. rho=-0.69, p<0.001
3. rho=-0.68, p<0.001
Tripathy 2014[27] / Criterion predictive validity Katz ADL / 1. 28-day mortality
3. Long-term mortality / 1. OR[95%CI]=0.20 [0.10-0.80], p=0.03
2. NS association (p=0.06)
Baldwin 2014[28] / Criterion predictive and construct validity FFI / 1. 1-month disability (Katz ADL)
2. 6-month mortality / 1. HR[95%CI]=2.20 [1.04-4.60]
2. HR[95%CI]=3.00 [1.40-6.30]
Nordon-Craft 2014 [30] / Construct and criterion predictive validity PFIT-s / 1. MRC-SS
2. Grip strength
3. Predictive utility
4. MRC-SS predictive of standing components of PFIT-s / 1. rho=0.923, p<0.005
2. rho=0.763, p<0.005
3. Higher PFIT-s scores predictive of reduced likelihood of discharge to LTACH at baseline: OR [95%CI]=0.63 [0.40-0.99], p-0.043 and ICU discharge: OR [95%CI]=0.70 [0.51-0.98], p=0.038
At baseline and ICU discharge PFIT-s scores not predictive of discharge to home (p=0.087)
4. ICU DC MRC-SS cut point of 41.5 out of 60 predictive of ability to perform standing components of the PFIT-s
Denehy 2013[32] / Criterion predictive and construct validity PFIT (at ICU DC) / 1. 6MWT
2. TUG test
3. MRC
4. BMI
5. Predictive utility / 1. r[95%CI]=0.41 [0.24-0.55], p<0.001
2. r[95%CI]=-0.60 [-0.70 -0.46}, p<0.001
3. rho 95%CI]=0.49 [0.33-0.62], p<0.001
4. ICU admission PFIT r [95%CI]=-0.01 [-0.18-0.16]
5. Higher PFIT predictive of higher MRC (OR=1.28, p<0.001); DC home (OR=1.20, p=0.001) and reduced likelihood of DC to rehab (OR=0.86, p=0.02)
Higher admission PFIT was associated with reduced acute hospital LOS (B coef=-2.13, p< 0.001)
Admission PFIT was not predictive for ICU or hospital readmission, mortality or SF-36 PF score at 3, 6 or 12 month follow up
Corner 2013[33] / Construct validity CPAx / 1. Peak cough flow
2. AusTOMs MSK score (activity)
3. AusTOMs MSK score impairment
4. MRC
5. MV days
6. SF-36 PF domain
7. SF-36 MF domain / 1. r=0.63, p=0.006; 2. r=0.74, p<0.001
3. r=0.76, p<0.001; 4. r=0.65, p<0.001
5. r=-0.51, p<0.001
6. r=0.72, p=0.01 for preadmission CPAx score and r=0.84, p=0.009 for ICU DC score
7. r=0.12, p=0.72 for preadmission CPAx score and r=0.02, p=0.95 for ICU DC score
Abd –El-Gaward 2013 [34] / Criterion predictive validity Katz ADL /
- 30-day mortality
- MV duration
2. AUC=0.68, p=0.02; sensitivity=96%; specificity=40%; PPV=53% and NPV=94%
Thrush 2012 [35] / Construct validity FSS-ICU / 1. DC destination / 1. Significant differences between discharge FSS-ICU score of the five discharge settings (Chi square=43.11, p<0.001)
Kasotakis 2012 [36] / Construct and Criterion predictive validity SOMS / 1. ICU LOS
2. Hospital LOS
3. In-hospital mortality
4. HGD
5. HGD and above 1-3 items / 1. r=-0.54, p<0.001
2. r=-0.37, p<0.001
3. r=-0.33, p=0.001
4. r=0.51, p=0.008
5. HGD not predictive of outcomes p0.05
Allison 2012 [37] / Construct validity 6MWT / 1. SF-36 PF domain / 1. r=0.62 (week 1), r=0.55 (week 8) and r=0.47 (week 26), p<0.001
Vest 2011 [38] / Construct validity Katz ADL / 1. SF-36 PCS
2. SF-35 MCS / 1. After adjusting for covariates at one month, ß=-7.11; p<0.001 and at 12-months NS (p=0.13)
2. NS at one and 12-months (p>0.05)
Daubin 2011 [39] / Criterion predictive validity Katz ADL / 1. 3-month mortality / 1. NS p=0.36
Clini 2011 [40] / Construct and criterion predictive validity admission for change in Katz ADL / 1. Baseline Katz ADL
2. Baseline FIM
3. Baseline quads
4. Baseline biceps
5. Probability of being ventilator free and survival / 1. r=0.16, p=0.13
2. r=-0.06, p=0.34
3. r=0.27, p=0.03
4. r=0.24, p=0.05
5. Probability of remaining ventilator free (p=0.043) and survival (p=0.001) differed across change in Katz ADL categories
Sacanella 2009[42] / Criterion predictive validity Barthel and Lawton IADL / 1. In-hospital mortality
2. Post hospital mortality
3. Cumulative mortality / 1. NS association (p>0.05)
2. Barthel NS (p>0.05), Lawton p=0.001 on univariate analysis
3. Barthel (p=0.03), Lawton p=0.007 on univariate analysis
Patients with baseline Lawton index<5; HR[95%CI]=4.10 [1.50-11.00], p=0.006
Van der Schaaf 2008 [43] / Construct validity Barthel / 1. FAC
2. HGD
3. MV duration / 1. r=0.76, p=0.000
2. r=0.62, p=0.000
3. r -0.474, p=0.001
Chiang 2006[45] / Construct validity change in Barthel score and FIM score (at week 6 from baseline) / Change from week 6 to baseline for:
1. PiMax
2. Shoulder flexor strength
3. elbow flexor strength
4. knee extensor strength
5. ventilator free time / Barthel
1. r=0.67, p<0.05
2. r=0.70, p<0.05
3. r=0.83, p<0.05
4. r=0.68, p<0.05
5. r=0.65, p<0.05
FIM
1. r=0.64, p<0.05
2. r=0.62, p<0.05
3. r=0.71, p<0.05
4. r=0.60, p<0.05
5. r=0.69, p<0.05
Bo 2003 [46] / Criterion predictive validity Katz ADL and Lawton IADL / 1. In-hospital mortality / 1. Lawton IADL and Katz ADL significantly associated p<0.001
Katz ADL1 function lost; OR[95%CI]=5.10 [3.20-8.00]; Lawton IADL 5 functions autonomous OR[95%CI]=4.90[2.80-8.60]
Dardaine 2001 [47] / Criterion predictive validity MAC and Katz ADL / 1. ICU mortality
2. 6-month mortality / 1. MAC NS (p>0.05); Katz ADL NS (p>0.05)
2. MAC under 10th percentile (p≤0.001) on Univariate; MAC OR[95%CI]=3.43[1.10-0.60] p≤0.05; Katz ADL: NS (p>0.05)
Abbreviations: ACM/BW ratio; active cell mass / body weight ratio; ADL, activities of daily living; AUC, area under the curve; AusTOMs, Australian Therapy outcome Measures; BMI, body mass index; BIS, bioimpedance spectroscopy; CPAx, Chelsea Critical Care Physical Assessment tool; D, day; DC, discharge; Dx, diagnosis; ECW/ICW ratio, extracellular water / intracellular water ratio; ECM, extracellular mass; EMS, electrical muscle stimulation; FAC, functional ambulation categories; FIM, functional independence measure; FFI, Fried’s frailty index; FFM, fat-free mass; FSS-ICU, functional status score in intensive care; HGD, hand-grip dynamometry; HR, hazard ratio; IADL, instrumental activities of daily living; ICU, intensive care unit; ICU-AW, intensive care unit acquired weakness; IRR, incidence rate ratio; HGD, handgrip dynamometry; KUH, Kansas University Hospital; L, left; LOS, length of stay; MAC, midarm circumference; MCS, mental component summary score; MF, mental function; MIP, maximal inspiratory pressure; MRC, Medical Research Council; MSK, musculoskeletal; MV, mechanical ventilation; NPV, negative predictive value; NS, non significant; O2, oxygen; OR, odds ratio; PCS, physical component summary score; PF, physical function; PFIT, physical function in intensive care test; PiMax, Peak Inspiratory Maximal Pressure; PPV, positive predictive value; quads, quadriceps; r, correlation; R, right; SD, standard deviation; SOMS, surgical optimal mobility scale; SF-36, Short Form 36 health survey; TUGT, timed up and go test; TISS-28, Therapeutic index scoring system 28; US, ultrasound; vs, versus; %, percentage; 6MWT, six-minute walk test; 95%CI, ninety-five percent confidence interval.