Isokinetic testing of muscle strength of older individuals post-stroke: An integrative review

Bohannon, Richard W.

doi:10.3233/IES-201146

Isokinetic testing of muscle strength of older individuals post-stroke: An integrative review

Issue title: Aging and Isokinetics

Guest editors: Richard Bohannon

Article type: Review Article

Authors: Bohannon, Richard W.

Affiliations: Department of Physical Therapy, College of Pharmacy and Health Sciences, Campbell University, Lillington, NC 27546, USA | Tel.: +1 910 814 4096; E-mail: bohannon@campbell.edu

Correspondence: [*] Corresponding author: Department of Physical Therapy, College of Pharmacy and Health Sciences, Campbell University, Lillington, NC 27546, USA. Tel.: +1 910 814 4096; E-mail: bohannon@campbell.edu.

Keywords: Muscle strength, measurement, clinimetrics

DOI: 10.3233/IES-201146

Journal: Isokinetics and Exercise Science, vol. 28, no. 3, pp. 303-316, 2020

Published: 31 July 2020

Get PDF

Abstract

BACKGROUND:

Muscle weakness is a common impairment accompanying stroke.

OBJECTIVE:

Describe isokinetic testing procedures and clinimetric findings associated with the testing of older individuals with weakness following stroke.

METHODS:

Relevant articles were identified by an electronic PubMed search using the search string “isokinet* AND stroke.”

RESULTS:

Seventy-six relevant articles were found. The articles largely support the validity and reliability of isokinetic strength testing of older patients with stroke. Little data are available that provides specific information on the responsiveness for such measures.

CONCLUSIONS:

Isokinetic dynamometry is a valid and reliable measure of muscle strength after stroke. However, it is not particularly practical and information is lacking regarding its responsiveness.

1.Introduction

Stroke is the most common neurological disorder in the world today. The World Stroke Organization has estimated that the worldwide incidence of stroke is 15 million per year and that stroke is the second leading cause of death for adults older than 60 years [1]. Although numerous impairments can result directly from stroke, muscle weakness is probably the most common and obvious [2]. Several options are available for measuring muscle strength after stroke, among them are self-report [3] manual muscle testing [4], field tests such as sit-to-stand [5], hand-held dynamometry [6], hand-grip dynamometry [7], and isokinetic dynamometry [8]. This review focuses on isokinetic dynamometry. Specifically, it seeks to describe the isokinetic testing procedures that have been applied to older individuals with stroke and the findings obtained using the procedures.

2.Methods

Potentially relevant articles were identified by a search of PubMed on February 10, 2019. The search string used was “isokinet* AND stroke.” A hand-search was also conducted. Article titles and abstracts, and where warranted full text, were examined to determine whether articles identified by the searches addressed the isokinetic testing of muscle strength of individuals with stroke. Review articles and articles in languages other than English were excluded. Articles were also excluded if they failed to provide procedural specifics (e.g., dynamometer used), used a custom isokinetic dynamometer, or presented no findings relevant to the clinimetric properties of isokinetic testing. Articles retained for this review were mined for information on the sample tested, procedural specifics (e.g., muscle groups tested and speed of testing), and clinimetrically relevant findings (e.g., information on validity, reliability, and responsiveness).

Table 1

Summary of studies describing the isokinetic measurement of strength after stroke

Study	Participants	Procedures	Findings
Abdollahi et al. (2015) [8]	Iranians with chronic stroke (n= 40, mean age = 59.3 y)	Cybex Model 770 dynamometer measured concentric strength (PT normalized against nonparetic side & bodyweight) of knee extension @ 90∘/s & ankle plantarflexion @ 30∘/s.	Validity: paretic side strength normalized against nonparetic side correlated SGNF (τ-b ⩾ 0.36) with Modified Ashworth, but paretic side strength normalized against weight was not correlated SGNF (τ-b ⩽-0.20) with Modified Ashworth. Reliability: test-retest (1 wk apart) – knee extension strength ICC = 0.94 (weight normalized) & ICC = 0.86 (nonparetic side normalized); ankle plantarflexion strength ICC = 0.80 (weight normalized) & ICC = 0.70 (nonparetic side normalized).
An and Jo (2017) [9]	Koreans with chronic stroke assigned to ankle mobilization group (n= 13, mean age = 48.5y) or control group (n= 13, mean age = 48.3 y)	Biodex dynamometer measured B⃝ concentric paretic ankle dorsiflexion & plantarflexion strength (PT normalized against body weight) @ 30∘/s.	Responsiveness: ankle strength ↑ SGNF in the mobilization group but not the control group.
Andersen et al. (2011) [10]	Danish with chronic stroke (n= 11, mean age = 51.0 y)	KinCom dynamometer measured B⃝ knee flexion & extension strength (normalized average torque) concentrically @ 30 & 240∘/s & eccentrically @ 30∘/s.	Validity: paretic side strength < nonparetic side strength. Responsiveness: 12 weeks of intensive rehabilitation associated with SGNF ↑ in all normalized paretic strength measures.
Avila et al. (2013) [11]	Brazilians with chronic stroke (n= 22, mean age = 65.5 y) & healthy controls (n= 22, mean age = 64.4 y)	Biodex Multi-joint System 3 dynamometer measured B⃝ concentric shoulder abduction strength (PT, total work, average power) @ 60∘/s.	Validity: all paretic side strength measures < nonparetic side measures, torque & work of paretic side SGNF < nonparetic side. All strength measures of the nonparetic and paretic sides < controls with all paretic side measures SIGN < controls and nonparetic work & power SGNF < controls.
Barbic and Brouwer (2008) [12]	Canadians with chronic stroke (n= 10, mean age = 60.6 y), healthy old group (n= 10, mean age = 62.1 y) & healthy young group (n= 10, mean age = 20.7 y) controls	Biodex System 3 dynamometer measured B⃝ concentric hip flexion & extension strength (gravity normalized PT) @ 60∘/s.	Validity: strength of patients with stroke < controls & strength of paretic side < that of nonparetic side.
Bohannon (1992) [13]	Americans with subacute stroke (n= 20, mean age = 63.7 y)	Lido Active Rehabilitation System measured B⃝ knee extension strength (maximum velocity, isometric PT & power). Speed was set @ 400∘/s for velocity & power. Isometric torque was measured at 90∘.	Validity: strength of paretic side SGNF < nonparetic side. All strength measures were correlated SGNF with comfortable and maximum gait speed, but correlations with paretic side measures (r= 0.63 to 0.75) > nonparetic side measures (r= 0.45 to 0.65).
Bohannon (1987) [14]	Americans with subacute stroke (n= 27, mean age = 64.1 y)	Cybex II measured B⃝ knee extension strength (torque) @ 30, 60, 120, & 189∘/s.	Validity: strength of paretic side < that of nonpatretic side at every speed. Strength correlated SGNF (r= 0.74 to 0.94) between measurements @ different speeds of paretic and nonparetic sides.
Brogårdh et al. (2012) [15]	Swedes with chronic stroke assigned to vibration group (n= 16, mean age = 61.3 y) or control group (n= 15, mean age = 63.9 y)	Biodex Multi-Joint System 3 PRO dynamometer measured B⃝ isokinetic concentric knee extension & flexion strength (PT) @ 60∘/s & isometric knee extension strength (peak torque) @ 90∘.	Validity: strength of paretic side < that of nonparetic side.
Calmels et al. (2011) [16]	French with subacute or chronic stroke (n= 14, mean age = 53.7 y)	Cybex 6000 dynamometer measured B⃝ knee extension strength (body weight normalized PT) concentrically @ 60 & 120∘/s & isometrically @ 30 & 60∘.	Validity: strength of paretic side < nonparetic side. Responsiveness: strength after ergometer training. ↑ SGNF in 5 of 8 nonparetic measures and 3 of 8 paretic measures.

Table 1, continued
Study	Participants	Procedures	Findings
Carvalho et al. (2013) [17]	Swedes with chronic stroke (n= 41, mean age = 59 y) & healthy reference group (n= 144, age range = 40–79 y)	Biodex Multi-Joint System 3 PRO dynamometer measured concentric B⃝ knee flexion & extension strength (maximum torque) @ 60∘/s & isometric ankle dorsiflexion and plantarflexion strength (torque) @ 0∘. A strength index was calculated using all 8 measures.	Validity: mean strength of paretic side of patients was < controls. Mean strength of nonparetic side < controls. Mean strength index for stroke group < that of control group. For patients with stroke, strength index correlated SGNF with self-selected gait speed (r2= 0.37), maximum gait speed (r2= 0.63) & 6MWT distance (r2= 0.44).
Chen et al. (2015) [18]	Taiwanese with subacute stroke assigned to isotonic training (n= 12, mean age = 67.1 y) or isokinetic training (n= 12, mean age = 64.7 y)	Biodex dynamometer measured B⃝ isometric knee flexion & extension strength (PT) @ 90∘ & isokinetic knee flexion & extension strength (PT) @ 60 & 120∘/s.	Responsiveness: knee strength ↑ in both groups after strength training, ↑ in isokinetic group were all SGNF.
Clark et al. (2006) [19]	Americans with chronic stroke (n= 17, mean age = 57.5 y) & nondisabled controls (n= 13, mean age = 63 y)	Biodex System 3 Pro dynamometer measured paretic knee extension strength (torque & power) isometrically @ 0∘ concentrically @ 30, 60, 90, 120, 150, 180, 210, & 240∘/s & eccentrically @ 30, 60, 90, 120, 150 & 180∘/s.	Validity: knee extension strength (as % of maximum capacity) of paretic knee extensors < controls. Reliability: test-retest 48 hr apart – ICC (torque) = 0.86 to 0.97 & ICC (power) = 0.83 to 0.97. Responsiveness: smallest real difference (%) 39.0 to 72.7 (torque) & 38.7 to 78.1 (power).
Cohen et al. (2018) [20]	Canadians with subacute stroke (n= 75, mean age = 61.1 y)	Biodex System 3 Pro dynamometer measured B⃝ concentric knee & hip flexion & extension strength (PT & power) @ 60∘/s & ankle dorsiflexion and plantarflexion strength (PT & power) @ 30∘/s on both sides. Strength measures of each side were summed & normalized against body weight.	Validity: strength of the nonparetic lower limb was SGNF > strength of the paretic lower limb. Strength measures were correlated SGNF with the physical component of the SIPSO (r= 0.42 to 0.59) & the physical component summary of the SF-36 (r= 0.29 to 0.47).
Coroian et al. (2018) [21]	French with chronic stroke assigned to isokinetic strengthening (n= 10, mean age = 63.6 y) or passive mobilization (n= 10, mean age = 63.6 y)	CON-TREX dynamometer measured elbow flexor & extensor strength (PT) @ 30∘/s & wrist flexor & extensor strength (PT) @ 15∘/s.	Responsiveness: strength ↑ over 45 days in wrist & elbow muscle groups (effect sizes 0.11 to 0.44).
Dehkordi et al. (2008) [22]	Iranians with chronic stroke (n= 42, mean age = 61.5 y)	Biodex dynamometer measured B⃝ concentric knee flexion & extension strength (PT & normalized PT) @ 60 & 120∘.	Validity: strength of the affected side SGNF < unimpaired side. Reliability: test-retest (1 day apart) – ICC (torque) = 0.88 to 0.99 & ICC (normalized torque) = 0.85 to 0.98. Responsiveness: smallest real difference (%) 18.7 to 85.1 (torque) & 22.3 to 76.3 (normalized torque).
Dias et al. (2017) [23]	Brazilians with chronic stroke (n= 15, mean age = 56 y) or healthy controls (n= 15, mean age = 59 y)	Biodex dynamometer measured isometric ankle plantar flexion strength (PT normalized against body mass) at maximum dorsiflexion, 0∘ & 30∘.	Validity: strength of affected limb SGNF < unaffected limb which was SGNF < control limb.
Ekiz et al. (2015) [24]	Turks with subacute/ chonic stroke assigned to kinesio tape (n= 12, mean age = 48.8 y) or control group (n= 12, mean age = 50.9 y)	Biodex System 3 Pro Multijoint System dynamometer measured B⃝ knee flexion and extension strength (PT) @ 60 & 180∘/s.	Responsiveness: after 4 weeks of training, paretic side & nonparetic side strength ↑ SGNF in all measures in both groups.

Table 1, continued
Study	Participants	Procedures	Findings
Ekstrand et al. (2015) [25]	Swedes with chronic stroke (n= 45, mean age = 65 y)	Biodex System 3 PRO dynamometer measured B⃝ shoulder isometric abduction strength (torque) @ 15∘ of shoulder abduction, isometric elbow flexion strength (torque) @ 30∘ of elbow flexion), & isokinetic elbow flexion & extension strength (PT) @ 60∘/s.	Reliability: test-retest (1 week apart) – ICC (less affected/more affected) isometric shoulder abduction = 0.97/0.97, isometric elbow flexion = 0.97/0.97, isokinetic elbow flexion = 0.95/0.95, isokinetic elbow extension = 0.92/0.92. Responsiveness: SRD% (less affected/more affected) isometric shoulder abduction = 16.0/26.1, isometric elbow flexion = 15.5/21.2, isokinetic elbow flexion = 20.6/25.5 isokinetic elbow extension = 25.9/34.8.
Ekstrand et al. (2016) [26]	Swedes with chronic stroke (n= 45, mean age = 65 y)	Biodex System 3 PRO dynamometer measured B⃝ isometric shoulder abduction @ 15∘ & elbow flexion @ 90∘ strength (PT) & isokinetic elbow flexion & extension strength (PT) @ 60∘/s.	Validity: grip strength was correlated SGNF with other strength measures of both sides: isometric shoulder abduction (r= 0.82 & 0.80) & elbow flexion (r= 0.77 & 0.82) strength & isokinetic elbow flexion (r= 0.76 & 0.81) & elbow extension (r= 0.65 & 0.77).
Eng et al. (2002) [27]	Canadians with chronic stroke (n= 20, mean age = 60 y)	Kim-Com dynamometer measured B⃝ concentric hip, knee, & ankle flexion & extension strength (PT & average torque) @ 60∘/s.	Validity: all strength measures of non-involved side SGNF > paretic side. Reliability: test-retest (2–4 days apart) – ICC (PT) = 0.97 to 0.99 paretic side & 0.95–0.98 non-involved side. ICC average torque = 0.96 to 0.98 paretic side & 0.88 to 0.96 non-involved side.
Eng et al. (2009) [28]	Canadians with chronic stroke (n= 18, mean age = 64.9 y) & neurologically healthy (n= 18, mean age = 63.3 y)	Kin-Com dynamometer measured B⃝ concentric & eccentric hip, knee, & ankle flexor & extensor strength (PT) @ 30∘/s.	Validity: strength of controls > strokes and strength of nonparetic side > paretic side of strokes. Physical activity correlated SGNF with concentric torque preservation (r= 0.51 & 0.64) but not eccentric torque preservation.
Engardt et al. (1995) [29]	Swedes ambulatory after stroke assigned to concentric training (n= 10, mean age = 64.6 y) or eccentric training (n= 10, mean age = 62.2 y)	Kin-Com 500H dynamometer measured B⃝ concentric and eccentric strength (average torque) @ 60, 120, & 180∘/s.	Responsiveness: concentric & eccentric strength ↑ SGNF in the concentric & eccentric training groups.
Fernandez-Gonzalo et al. (2014) [30]	Swedes with chronic stroke (n= 12, mean age = 63.3 y)	IsoMed 2000 dynamo-meter measured B⃝ concentric and eccentric knee extension strength (PT over 1 s window) @ 30, 60, & 90∘/s.	Validity: all strength measures of affected limb SGNF < nonaffected limb. Responsiveness: all strength measures of the affected knee ↑ after 8 weeks of flywheel resistance training. Some (4 of 7) strength measures of the nonaffected knee ↑ after training.
Flansbjer et al. (2006) [31]	Swedes with chronic stroke (n= 50, mean age = 58 y)	Biodex Multi-joint System II dynamometer measured B⃝ concentric knee extension & flexion strength (PT) @ 60∘/s.	Validity: strength of paretic knee extension & flexion < nonparetic knee extension & flexion. Nonparetic knee extension & flexion strength for men was SGNF > for women. Paretic knee extension strength for men was SGNF > for women. Paretic knee flexion strength was not SGNF different for men and women. Paretic knee extension strength, unlike nonparetic knee extension strength, was correlated SGNF with Timed up & Go (r=-0.65), comfortable gait speed (r= 0.61, fast gait speed (r= 0.67), stair climb ascent (r=-0.58), stair climb descent (r=-0.61), & 6 min walk test (r= 0.70). Paretic knee flexion strength, unlike nonparetic knee flexion strength, was correlated SGNF with timed up & go (r=-0.64), comfortable gait speed (r= 0.61, fast gait speed (r= 0.65), stair climb ascent (r=-0.61), stair climb descent (r=-0.61), & 6 min walk test (r= 0.71).

Table 1, continued
Study	Participants	Procedures	Findings
Flansbjer et al. (2005) [32]	Swedes with chronic stroke (n= 50, mean age = 58 y)	Biodex Multijoint System II dynamometer measured B⃝ knee strength (torque) at 60∘ & 120∘ concentrically & 60∘ eccentrically.	Reliability: test-retest 7 days apart – ICC = 0.89 to 0.96. Responsiveness: smallest real difference 22–55%.
Flansbjer et al. (2012) [33]	Swedes with chronic stroke (mean age = 66 y) assigned to strength training group (n= 11) or control group (n= 7)	Biodex Multijoint System 3 PRO dynamometer measured B⃝ knee flexion & extension strength (torque) @ 60∘/s.	Validity: strength of nonparetic side SGNF > that of paretic side. Responsiveness: strength ↑ after strength training.
Flansbjer et al. (2008) [34]	Swedes with chronic stroke assigned to resistance training (n= 15, mean age = 61 y) or usual activity (n= 9, mean age = 60 y)	Biodex Multi-Joint System 3 PRO dynamometer measured B⃝ concentric knee flexion & extension strength (PT) @ 60∘/s.	Validity: strength of paretic knee muscles < that of nonparteic knee muscles. Responsiveness: strength ↑ over course of intervention in training group.
Freire et al. (2017) [35]	Brazilians with chronic stroke (n= 12, mean age = 59.6 y) or healthy controls (n= 12, mean age = 59.8 y)	Biodex Medical System 3 dynamometer measured B⃝ isometric ankle plantarflexion strength (normalized torque) @ 0∘ & maximum dorsiflexion.	Validity: maximum plantarflexion strength and strength @ 0∘ SGNF < on the paretic than nonparetic side & on nonparetic side than healthy controls.
Gray et al. (2017) [36]	Americans with chronic stroke (n= 18, mean age = 61.4 y)	Biodex System Pro4 dynamometer measured B⃝ isokinetic ankle dorsiflexion & plantarflexion strength & hip abduction & adduction strength (PT) @ 30∘/s.	Validity: all strength measures of paretic limb < nonparetic limb.
Hameau et al. (2014) [37]	French with chronic stroke (n= 14, mean age = 54 y)	ConTrex – MJ dynamometer measured paretic isokinetic knee flexion & extension strength (PT) @ 30, 60, & 90∘/s & isometric knee flexion & extension strength (PT) @ 40∘ & 60∘.	Responsiveness: all strength measures except knee flexion @ 90∘/s ↑ SGNF after botulinum toxin injection.
Hamrin et al. (1982) [38]	Swedes with chronic stroke assigned to special activation (n= 23, mean age = 69.9 y) or routine activation (n= 14, mean age = 69.7 y) or healthy controls (n= 13, age = not stated)	Cybex II dynamometer measured knee extension & flexion strength & elbow extension and flexion strength (torque) @ 45 & 60∘ during knee testing @ 30 & 90∘/s & @ 90∘ during elbow testing @ 30 & 90∘/s.	Validity: paretic limb strength < nonparetic limb strength. Both paretic and nonparetic limb strength < that of controls. All paretic knee strengths correlated SGNF with locomotion (r= 0.71 to 0.90). All but 2 nonparetic knee strengths were correlated SGNF with locomotion (r= 0.47 to 0.67). Of 48 correlations between elbow strength and 3 activities (e.g., dressing), 8 involving paretic elbow muscles were significant (r= 0.48 to 0.77) & 2 involving nonparetic elbow muscles were significant (r= 0.63 & 0.68).
Hsu et al. (2003) [39]	Taiwanese with chronic stroke (n= 26, mean age = 54.2 y)	Cybex 6000 dynamometer measured B⃝ hip flexor, knee extensor, & ankle plantarflexor strength (normalized PT & total work) @ 30∘/s (hip & ankle), 90∘/s (knee).	Validity: affected limb strength < unaffected limb strength. All affected side strength measures correlated SGNF with comfortable speed (r= 0.39 to 0.57) and fast speed (r= 0.40 to 0.71) gait. Affected side strength measures correlated SGNF with some measures of gait asymmetry (r=-0.25 to -0.57).
Hsu et al. (2002) [40]	Taiwanese with chronic stroke (n= 9, mean age = 55.6 y)	Cybex 6000 dynamometer measured B⃝ hip flexor, knee extensor, & ankle plantarflexor strength (normalized PT, total work & average power) @ 30∘/s (all 3 muscle groups), 90∘/s (hip flexors & knee extensors) & 15∘/s (ankle plantarflexors).	Validity: affected limb strength < unaffected limb strength, usually SGNF. Reliability: test-retest 1 week apart – ICC = 0.62 to 0.96. ICC for strength deficits = 0.13 to 0.91.

Table 1, continued
Study	Participants	Procedures	Findings
Hyun et al. (2015) [41]	Koreans with subacute stroke (n= 53, mean age = 62.6 y)	Humac Norm measured B⃝ isometric knee flexion & extension strength (gravity-corrected PT) @ 60∘ of knee flexion.	Validity: paretic limb strength < nonparetic limb strength. Strength of household ambulators SGNF < community ambulators. Paretic knee extensor strength related to 6 min walk test (adjusted R2= 0.67).
Karatas et al. (2004) [42]	Turks with acute stroke (n= 38, mean age = 59.1 y) or healthy controls (n= 40, mean age = 62.6 y)	Cybex 770 NORM dynamometer measured concentric & isometric trunk flexion & extension strength (PT) @ 60, 90, 120 & 0∘/s. Isometric strength measured at trunk flexion angle of 60∘.	Validity: strength measures of healthy controls > patients. Strength measures SGNF higher except for isometric flexion & extension & concentric extension @ 120∘/s. Of 64 correlations between trunk strength & Functional Independence scores (r= 0.01 to 0.51) and Berg Balance scores (r= 0.10 to 0.64), 26 were SGNF.
Kim and Eng (2003) [43]	Canadians with chronic stroke (n= 20, mean age = 61.2 y)	Kin-Com measured B⃝ hip, knee, and ankle flexors & extensor strength (average torque normalized to body mass) @ 60∘/s.	Validity: paretic side strength < nonparetic side strength. Of 6 correlations between paretic side strength & self-selected gait speed (r= 0.33 to 0.84), 3 were significant. Of 6 correlations between nonparetic side strength & self-selected gait speed (r= 0.29 to 0.62), 2 were significant. Of 6 correlations between paretic side strength & stair speed (r= 0.27 to 0.71), 3 were significant. Of 6 correlations between nonparetic side strength & stair speed (r= 0.29 to 0.48), 2 were significant.
Kim et al. (2001) [44]	Canadians with chronic stroke assigned to isokinetic strength training (n= 10, mean age = 60.4 y) or passive range of motion (n= 10, mean age = 61.9 y)	Kin-Com dynamometer measured B⃝ hip flexion & extension, knee flexion & extension, & ankle dorsiflexion & plantarflexion strength (average torque normalized against body mass) @ 60∘/s.	Validity: strength of paretic side < that of nonparetic side. Responsivenes: composite strength of training group ↑ over course of intervention.
Kim et al. (2005) [45]	Americans with chronic stroke (n= 10, mean age = 62.2 y)	Biodex System 3.0 measured B⃝ elbow flexion & extension & shoulder flexion strength (torque, speed, power) at 30, 75 & 120∘/s.	Validity: strength of paretic side < nonparetic side. Reliability: test-retest session 1, day 7, week 6 – ICC = 0.73 to 0.99 on nonparetic side & 0.64 to 0.99 on the paretic side.
Kim et al. (2016) [46]	Koreans with chronic stroke assigned to ankle biofeedback training (n= 14, mean age = 51.2 y) or ankle strengthening (n= 13, mean age = 52.8 y)	Biodex dynamometer measured isometric affected ankle dorsiflexion & plantar flexion strength (PT) with ankle @ 60∘.	Responsiveness: strength ↑ SGNF in both biofeedback & ankle strengthening groups.
Knorr et al. (2010) [47]	Canadians with chronic stroke (n= 44, mean age = 62.6 y)	Biodex dynamometer measured B⃝ concentric hip & knee flexion & extension strength (body-weight normalized PT) @ 60∘/s & ankle dorsiflexion and plantarflexion strength (PTue) @ 30∘/s.	Validity: strength of paretic lower limb SGNF < strength of nonparetic lower limb. Strength of paretic limb correlated SGNF with Community Balance & Mobility Scale, (rs= 0.67), Berg Balance Scale (rs= 0.50) & Timed Up & Go (rs=-0.71). Strength of nonparetic limb correlated, albeit not always SGNF, with Community Balance & Mobility Scale, (rs= 0.46), BBS (rs= 0.28) & Timed Up & Go (rs=-0.44).
Kwong et al. (2017) [48]	Chinese with chronic stroke (n= 105, mean age = 61.0 y)	Cybex 6000 dynamometer measured paretic concentric knee extensor and flexor strength (PT) @ 90∘/s.	Validity: strength measures correlated SGNF with one another (rp= 0.60) & with 6 minute walk test (rp= 0.40 & 0.46). Knee extension strength correlated SGNF with BBS (rp= 0.31) but knee flexor strength did not (rp= 0.25). Neither strength measure correlated significantly with the SIPSO (rp= 0.23 & 0.29).

Table 1, continued
Study	Participants	Procedures	Findings
Lau et al. (2012) [49]	Hong Kongese with chronic stroke assigned to whole body vibration (n= 41, mean age = 57.3 y) or control group (n= 41, mean age-57.4)	Cybex dynamometer measured paretic side isometric knee flexion & extension strength (body mass normalized PT) @ 70∘.	Responsiveness: strength ↑ SGNF but minimally over course of vibration training.
Lee et al. (2015) [50]	Koreans with stroke assigned to overground treadmill walking (n= 16, mean age = 50.9 y) or underwater treadmill walking (n= 16, mean age = 49 y)	Biodex dynamometer measured knee flexion & extension strength (PT) @ 60 & 120∘/s.	Responsiveness: After 6 wk of training all strength measures ↑ SGNF except knee flexion strength @ 120∘/s in overground group.
Lee and Kang (2013) [51]	Korean inpatients with stroke assigned to isokinetic eccentric strengthening (n= 10, mean age = 53.4y) or conventional therapy (n= 10, mean age = 53.9 y)	Cybex 770 dynamometer measured hip flexion & extension strength (PT) @ 90∘/s.	Responsiveness: after 6 wk the eccentric training group demonstrated a SGNF ↑ flexion & extension strength.
Lee et al. (2018) [52]	Koreans with subacute stroke assigned to hydrotherapy (n= 19, mean age = 57.6 y) or conventional rehabilitation (n= 18, mean age = 63.7 y)	Humac Norm dynamometer measured B⃝ isometric knee flexion & extension strength (PT) while knees were @ 60∘.	Responsiveness: ↑ in paretic knee flexion and extension strength SGNF greater in the aquatic training group (p< 0.05). ↑ in nonparetic knee flexion and extension strength were not SGNF different between groups.
Liao et al. (2016) [53]	Chinese with chronic stroke assigned to low intensity whole body vibration (n= 28, mean age = 60.8 y), high intensity whole body vibration (n= 28, mean age = 62.9 y), or control group (n= 28, mean age = 59.8 y)	Humac NormTM dynamometer measured B⃝ knee flexion & extension strength (body mass normalized torque) isometrically while knees were at 30 & 70∘, & concentrically & eccentrically @ 60∘/s.	Responsiveness: most strength measures stayed the same or increased. SGNF ↑ (without reference to group) were found in 5 of 16 strength measures.
Lindmark et al. (1995) [54]	Swedes with subacute stroke (n= 34, median age = 74 y men & 72 y women)	Cybex II dynamometer measured B⃝ knee flexion & extension strength (torque) @ 90 & 12∘/s.	Validity: strength of nonparetic limb > that of nonparetic limb, SGNF under most test conditions. Most strength measures correlated SGNF with free speed (r= 0.21 to 0.72) & fastest gait speed (r= 0.31 to 0.70).
Lindström et al. (1998) [55]	Swedes with stroke (n= 10, mean age = 40 y) & healthy controls (n= 22, mean age = 28 y)	Cybex II measured B⃝ knee flexion and extension strength (PT) @ 90∘.	Validity: strength of controls > that of patients with stroke. Strength of paretic side SIGN < nonparetic side.
Lomaglio and Eng (2005) [56]	Canadians with chronic stroke (n= 22, mean age 67.0 y)	Kin-Com measured B⃝ concentric strength (average body mass normalized torque) of ankle plantar flexion & dorsiflexion @ 30∘/s & knee & hip flexion & extension @ 60∘/s.	Validity: 3 of 6 strength measures of the paretic lower limb correlated SGNF with self-paced STS (r=-0.07 to -0.72) & fast-paced STS (r=-0.10 to -0.74). No measure of the nonparetic lower limb correlated SGNF with either STS measure.
Lum et al. (2004) [57]	Americans with chronic stroke (n= 14, mean age = 70.6 y)	Biodex System 3 Pro measured B⃝ elbow flexion & extension strength (isokinetic torque normalized against isometric torque) isometrically @ 90∘ & concentrically @ 30, 75, & 120∘/s.	Validity: strength of paretic limb SGNF < that of unaffected limb.

Table 1, continued
Study	Participants	Procedures	Findings
MacIntyre et al. (2010) [58]	Canadians with subacute stroke (n= 11, mean age = 69 y), chronic stroke (n= 11, mean age = 72 y), or controls (n= 11, mean age = 71 y)	Biodex System 3 measured knee extension and ankle plantarflexion strength (body mass normalized torque) at 60 & 30∘/s respectively.	Validity: strength of patients with stroke SGNF < controls.
Maeshima et al. (2003) [59]	Japanese with stroke assigned to conventional training (n= 18, mean age = 64.8 y) or self-training with family support (n= 42, mean = 65.8 y)	Biodex System 3 measured nonparetic knee extension and flexion strength (body mass normalized PT) @ 60∘/s & 180∘/s.	Validity: strength of ambulatory patients SGNF > nonambulatory patients.
Marigold et al. (2004) [60]	Canadians with stroke (n= 40, mean age = 67.1 y) or controls (n= 60, mean age = 67.1 y)	Kin-Com dynamometer measured B⃝ concentric strength (PT) of the ankle flexors & extensors (30∘/s) & knee & hip flexors & extensors (60∘/s).	Validity: Muscle strength had SGNF relationships (r= 0.33 to 0.53) with only “the most challenging” sensory organization test items.
Marque et al. (1997) [61]	French with acute stroke (n= 15, mean age = 65.6 y) or sedentary controls (n= 16, mean age = 65 y)	Cybex 600 measured non paretic flexion & extension strength of hip @ 60∘/s, wrist @ 120∘/s, elbow @ 90∘/s & eversion & inversion of the ankle @ 60∘/s.	Validity: nonparetic strength of patients was < controls, SGNF so @ hip, ankle, and wrist. Responsiveness: nonparetic strength ↑ SGNF between days 20 & 90 for hip flexion & extension, ankle inversion, elbow extension, & wrist flexion & extension.
Nakamura et al. (1985) [62]	Japanese with chronic stroke (n= 11, mean age = 53.8 y)	Cybex II dynamometer measured B⃝ extension strength (PT) isokinetically @ 30∘/s, 90∘/s & 180∘/s & isometrically at 30∘, 60∘, & 90∘.	Validity: affected side strength measures SGNF < unaffected side. All strength measures correlated SGNF with one another (r= 0.60 to 0.99). All paretic side strength measures except isometric (30∘) correlated SGNF with walking velocity and rate (r= 0.60 to 0.87). No nonparetic side strength measures correlated significantly with walking velocity or rate.
Nascimento et al. (2014) [63]	Brazilians with chronic stroke (n= 12, mean age = 52 y) & healthy controls (n= 12, mean age = 52 y)	Biodex Medical System 3 Pro dynamometer measured B⃝ shoulder internal & external rotation, shoulder flexion & extension, & scapular protraction & retraction strength (PT & work expressed as % of control group) @ 60∘/s.	Validity: all strength measures of patients with stroke < measures of controls and all measures of paretic side of patients with stroke < nonparetic side.
Novak and Brouwer (2012) [64]	Canadians with chronic stroke (n= 10, mean age = 60.1 y) & controls (n= 10, mean age = 59.4 y)	Biodex System 3 measured B⃝ concentric & eccentric strength (body mass normalized torque) of ankle plantarflexion @ 30∘/s, knee extension @ 60∘/s, and hip extension @ stair-climbing velocity.	Validity: strength of controls SGNF > patients with stroke. Concentric & eccentric strength of ankle plantarflexors & knee extensors of less affected side SGNF > more affected side.
Pang et al. (2013) [65]	Chinese with chronic stroke assigned to a whole body vibration group (n= 41, mean age = 57.3 y) or control group (n= 41, mean age = 57.4 y)	Cybex NUMAC dynamometer measured concentric and eccentric knee extension & flexion strength (peak power normalized to body mass) @ 60∘/s.	Responsiveness: most strength measures ↑, some SGNF.
Ploutz-Snyder et al. (2006) [66]	Americans with chronic stroke (n= 6, mean age = 55.9 y)	Biodex System 3 dynamometer measured B⃝ isometric elbow flexion and extension strength (torque) @ 90∘.	Validity: strength of affected side SGNF < that of unaffected side in both flexors & extensors.

Table 1, continued
Study	Participants	Procedures	Findings
Pohl et al. (2000) [67]	Americans with stroke (n= 10, mean age = 64.0 y) & controls (n= 10, mean = 68.7 y)	Cybex II dynamometer measured B⃝ knee extension & flexion strength (PT & average torque) @ 60∘/s & ankle extension & flexion strength (PT & average torque) @ 30∘/s.	Validity: affected side strength < unaffected side & < controls. Reliability: test-retest reliability (1 week apart) – ICC = 0.75 to 0.97 (less affected side) & 0.44 to 0.90 (more affected side).
Quintino et al. (2018) [68]	Brazilians with stroke (n= 18, mean age = 59.8 y) & healthy controls (n= 18, mean age = 57 y)	Biodex dynamometer measured concentric trunk flexion & extension strength (PT, torque @ 90∘, total work & total work normalized against trunk mass) @ 60∘/s & 120∘/s.	Validity: all strength measures of patients with stroke were SGNF < for healthy individuals.
Rabelo and Fachin-Martins (2018) [69]	Brazilians with chronic stroke (n= 13, mean age = 57.0 y) & able-bodied (n= 13, mean age = 63.0 y)	Biodex System 3 Pro measured trunk flexion and extension strength (PT, PT normalized to body weight, power & total work) @ 60∘/s & 120∘/s.	Reliability: inter-rater reliability – ICC = 0.61 to 0.96; test-retest reliability (1 week apart) – ICC = 0.84 to 0.98.
Ryan et al. (2011) [70]	Americans with chronic stroke (n= 70, mean age = 63 y)	Kin-Com 125AP measured B⃝ knee concentric & eccentric strength (PT) @ 90 & 120∘/s.	Validity: strength of paretic knee SGNF < nonparetic knee. 6MWT correlated SGNF with paretic (r= 0.55) & nonparetic (r= 0.33) eccentric strength & nonparetic concentric strength (r= 0.34).
Şen et al. (2015) [71]	Turks with subacute- chronic stroke assigned to isokinetic exercise (n= 25, mean age = 51.3 y) or control group (n= 25, mean age = 55.4 y) or & healthy individuals (n= 30, mean age = 49.9 y)	Biodex System 3 Pro Multijoint System dynamometer measured B⃝ knee extension and flexion strength (PT) @ 60∘/s & 180∘/s & ankle dorsiflexion and plantarflexion strength (PT) @ 60∘/s & 120∘/s.	Validity: strength of nonparetic strength measures all SGNF < control group. Responsiveness: strength of all paretic and nonparetic muscle groups (exercise & control) ↑ SGNF.
Sharp et al. (1997) [72]	Canadians with chronic stroke (n= 15, mean age = 67 y)	Cybex 2 dynamometer measured B⃝ knee extension and flexion strength (PT) @ 30, 60, & 120∘/s.	Validity: strength of the paretic side SGNF < nonparetic side. Responsiveness: strength ↑ SGNF with isokinetic training over time on paretic side.
Shimodozono et al. (2010) [73]	Japanese with subacute stroke assigned to isokinetic training + anabolic steroid group (n= 14, mean age = 66.6 y) or control group (n= 11, mean age = 62.3 y)	Cybex 6000 dynamometer measured nonparetic knee flexion & extension strength (bodyweight normalized PT) isometrically, isotonically & @ 60, 120 & 180∘/s.	Responsiveness: strength of nonparetic side ↑ over course of 6 week training with anabolic steroid.
Silva et al. (2015) [74]	Brazilians with stroke (n= 18, mean age = 59.8 y) & healthy controls (n= 18, mean age = 59.7 y)	Biodex dynamometer measured concentric trunk flexion & extension strength (PT & total normalized work) @ 60∘/s.	Validity: all strength measures of patients with stroke were SGNF < healthy individuals. Flexion strength correlated SGNF with 5 times STS test (r=-0.52 & -0.56). Extension strength correlated SGNF with 5 times STS test (r=-0.62 & -0.63).
Sunnerhagen and Mattsson (2005) [75]	Swedes with chronic stroke (n= 29, mean age = 53 y)	Kin-Com dynamometer measured B⃝ knee flexion & extension strength (torque) isometrically @ 60∘ & concentrically @ 60∘/s.	Validity: strength of paretic limb < nonparetic limb.
Suzuki et al. (1999) [76]	Japanese with subacute stroke (n= 34, mean age = 54.2 y)	Cybex II dynamometer measured B⃝ knee extension strength (PT) @ 30∘/sec.	Validity: strength correlated SGNF with most balance measures: sway path (r=-0.21 to -0.67), weight shifting (r= 0.37 to 0.73). Responsiveness: Strength ↑ SGNF over 4 & 8 week of gait training.

Table 1, continued
Study	Participants	Procedures	Findings
Suzuki et al. (1990) [77]	Japanese with stroke (n= 29, mean age = 57.8 y)	Cybex II dynamometer measured B⃝ knee extension strength (PT) @ 30∘/s.	Validity: strength of paretic side < nonparetic side. Strength of paretic side correlated SGNF with that nonparetic side. Strength of paretic side correlated SGNF with maximum walking speed (r= 0.85) & sway path (-0.63). Strength of nonparetic side correlated SGNF with maximum walking speed (r= 0.43) & sway path (r=-0.51)
Tanaka et al. (1997) [78]	Japanese men with stroke (n= 25, mean age = 61.3 y) or healthy controls (n= 25, mean age = 59.7 y)	Cybex trunk Extension-Flexion dynamometer measured trunk flexion & extension strength (PT) @ 0, 60, 120, & 150∘/s.	Validity: strength of men with stroke SGNF < controls except for flexion @ 0∘/s.
Tanaka et al. (1997) [79]	Japanese men with stroke (n= 65, mean age = 59.3 y males & 60.0 y females) & age-matched controls (n= 80, mean age = 60.9 y males & 59.3 y females).	Cybex Torso Rotation dynamometer measured left & right rotation strength (PT & best work) @ 60, 120, & 150∘/s.	Validity: strength of men SGNF > women. Strength of controls SGNF > individuals with stroke.
Tankisheva et al. (2014) [80]	Belgians with chronic stroke assigned to vibration training (n= 7, mean age = 57.4 y) or control (n= 8, mean age = 65.3 y) group	Biodex dynamometer measured B⃝ isometric knee flexion & extension strength (torque) at 60∘ & concentric knee flexion & extension strength (torque) @ 60∘/s & 240∘/s.	Responsiveness: all strength measures ↑, some SGNF, after 6 weeks vibration training.
Teixeira-Salmela et al. (1999) [81]	Canadians with chronic stroke assigned to exercise group (n= 6, mean age = 65.9 y) or control group (n= 7, mean age = 69.4 y)	Cybex II dynamometer measured hip & knee flexion & extension & ankle plantarflexion & dorsiflexion strength (total PT) @ 30 & 60∘/s.	Validity: torques generated at 2 speeds correlated SGNF (r= 0.77 to 0.98). Responsiveness: total PT of affected lower limb ↑ with training.
Tyson et al. (2013) [82]	British with stroke receiving electrical stimulation via a sock electrode (n= 29, mean age = 64.5 y)	Biodex dynamometer measured B⃝ dorsiflexion & plantarflexion strength (mean torque) @ 60∘/s.	Responsiveness: strength ↑ after stimulation with the ↑ SGNF for ankle plantarflexion.
Wang et al. (2014) [83]	Taiwanese with chronic stroke (n= 35, mean age = 57.1 y)	Biodex dynamometer measured B⃝ concentric knee extension strength (torque) @ 60∘/s & 90∘/s.	Validity: all strength measures correlated SGNF with walking speed (r= 0.45 to -0.66) and VO2 (r= 0.48 to 0.61).

*B⃝ = bilateral, PT = peak torque, SIPSO = Subjective Index of Physical and Social Outcome, SF-36 = Short Form Health Survey, SGNF = significant/significantly, BBS = Berg Balance Scale, r= Pearson correlation, rp= partial correlation, ICC = intraclass correlation coefficient. SRD = smallest real difference, τ-b = tau-b, STS = sit-to-stand.

3.Results and discussion

The PubMed search identified 161 potentially relevant articles. An additional 3 possibly relevant articles were identified by hand searches. Ultimately, 76 articles were found that met inclusion and exclusion criteria. Relevant information from those articles is summarized alphabetically by author in Table 1 [8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 70, 71, 72, 73, 74, 75, 76, 77, 78, 79, 80, 81, 82, 83].

The summarized information shows that isokinetic dynamometers have been used extensively to measure strength in individuals with stroke, most often older adults (mean/median age of 48.3 to 74.0 years) with chronic stroke. Isokinetic testing of individuals with stroke has been conducted in at least 15 different countries, but testing is reported most widely in North America and Europe. Utilization of some model of 7 different dynamometers has been described in the literature, with some type of Biodex described most frequently, followed by some model of Cybex or KinCom. Most articles that specified the type of measurement obtained documented concentric torque. Some articles however were inspecific or reported the measurement of eccentric or isometric torque with isokinetic dynamometers. A few articles reported torque normalized against body weight, the nonparetic side, or healthy controls [8, 9, 11, 16, 20, 22, 23, 35, 40, 43, 44, 53, 56, 58, 64, 65, 69, 73], or measures of work, power, or velocity [13, 40, 45, 63, 68, 69, 79]. The speed of concentric and eccentric testing ranged from 15 to 240∘/s, but testing was most commonly reported to be at 60∘/s [11, 12, 14, 16, 17, 18, 19, 20, 21, 22, 24, 25, 26, 27, 29, 30, 31, 32, 33, 34, 37, 42, 43, 44, 47, 50, 53, 56, 58, 60, 61, 62, 63, 64, 65, 66, 67, 68, 69, 71, 72, 73, 74, 75, 78, 79, 80, 81, 82, 83]. Testing at higher speeds was intentionally avoided in some studies as individuals with severe weakness following stroke can have difficulty producing torque at higher speeds on their paretic side [14, 57]. Most research reports bilateral isokinetic testing of limb muscles [9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 20, 22, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 39, 40, 41, 43, 44, 45, 47, 52, 53, 54, 55, 56, 57, 60, 62, 63, 64, 66, 67, 70, 71, 72, 75, 76, 77, 80, 81, 82, 83], particularly the knee flexors and extensors, but some articles limit testing to the paretic [19, 37] or nonparetic limb muscles. Additional research focuses on the testing of the trunk muscles [42, 78, 65, 79].

The clinimetric properties of isokinetic testing of individuals with stroke is generally well-supported. Validity has been demonstrated most often by differences in known groups and known conditions. Specifically, numerous studies have demonstrated that individuals with stroke are weaker, often significantly (SGNF), than controls in both the paretic [11, 17, 19, 28, 38, 55, 63, 64] and nonparetic [11, 17, 23, 28, 35, 38, 55, 61, 63, 71, 74, 78] limbs, that individuals with stroke are weaker (almost always SGNF) on their paretic side than on their nonparetic side [10, 13, 14, 15, 16, 20, 22, 23, 27, 28, 30, 31, 33, 34, 35, 36, 38, 40, 41, 43, 45, 47, 54, 55, 67, 68, 70, 72, 75, 77], that men tend to be SGNF stronger than women [79], that ambulatory individuals tend to be SGNF stronger than nonambulatory individuals [59], and that community ambulators are SGNF stronger than household ambulators [41]. Validity is also supported by correlations between isokinetically measured strengths of different muscle actions [26, 48] and between isokinetic strength measures at different speeds [14, 62, 81]. Most studies examining the correlation between isokinetic measurements of the lower limbs and mobility [13, 17, 31, 38, 39, 43, 47, 48, 54, 56, 62, 70, 74, 76, 77, 83] and other important variables [20, 60] have also reported SGNF results.

Numerous studies have used intraclass correlation coefficients (ICCs) to describe the relative reliability of strength measures obtained from individuals with stroke tested with isokinetic dynamometers. One study reported ICCs for inter-tester reliability of 0.61 to 0.96 [69]. All other studies addressing reliability focused on the test-retest consistency of measures obtained over periods of 1 day to 6 weeks. With the exception of a few results reported by Hsu et al. [40] and Kim et al. [45] virtually all test-retest ICCs reported exceeded 0.80 [8, 19, 22, 25, 27, 32, 40, 45, 67, 69]. The absolute reliability of isokinetic measures obtained from patients with stroke has received little attention. However, a few studies have reported a wide range of smallest real differences of 1.5 to 85.1% [19, 25, 15, 32], which also reflect on responsiveness. No studies were found that used minimal detectable changes or minimal clinically important differences to reflect responsiveness. What was found in abundance were studies showing a wide range of increases in isokinetic strength, some SGNF in response to strengthening and other interventions as well as the natural course of stroke recovery [11, 16, 18, 21, 24, 29, 30, 33, 34, 37, 44, 46, 49, 50, 51, 52, 53, 61, 65, 71, 72, 73, 76, 80, 81, 82]. Whether described using ICCs, smallest real difference or some other indication of reliability the magnitude of summary descriptions is dependent on the sensitivity of the isokinetic measurement, the time between measurements, the natural course of stroke recovery, and the effectiveness of any intervention applied between measurements.

In spite of precedence and evidence supporting the use of isokinetic testing of strength following stroke, the information in this nonsystematic review is limited. Most notably, a single bibliographic database (PubMed) was used. While it is doubtful that the inclusion of additional databases would have markedly altered the results, such expansion may have added to the evidence for the conclusions presented herein. A more systematic review may have also allowed for a meta-analysis of some variables and for a quality assessment of included articles.

4.Conclusion

There is considerable research support for measuring the isokinetic strength of older individuals who have experienced a stroke. That noted, the data are mostly limited to the young old and information on responsiveness is limited.

Conflict of interest

The author declares no conflict of interest.

References

[1]	Facts and figures about stroke. World Stroke Organization. https://www.world-stroke.org/component/content/article/16-forpatients/84-facts-and-figures-about-stroke. Accessed Februray 15, (2019) .
[2]	Lawrence ES, Coshall C, Dundas R, Stewart J, Rudd AG, Howard R, et al. Estimates of the prevalence of acute stroke impairments and disability in a multiethnic population. Stroke (2001) ; 32: : 1279-1284.
[3]	Duncan PW, Wallace D, Lai SM, Johnson D, Embretson S, Laster LJ. The Stoke Impact Scale Version 2.0. Evaluation of reliability, validity, and sensitivity to change. Stroke (1999) ; 30: : 2131-2140.
[4]	Demeurisse G, Demol O, Robaye E. Motor evaluation in vascular hemiplegia. Eur Neurol (1980) ; 19: : 382-389.
[5]	Mong Y, Teo TW, Ng SS. 5-repetition sit-to-stand test in subjects with chronic stroke: Reliability and validity. Arch Phys Med Rehabil (2010) ; 91: : 407-413.
[6]	Andrews AW, Bohannon RW. Distribution of muscle strength impairments following stroke. Clin Rehail (2000) ; 14: : 79-87.
[7]	Sunderland A, Tinson D, Bradley L, Hewer RL. Arm function after stroke. An evaluation of grip strength as a measure of recovery and a prognostic indicator. J Neurol Neurosug Psychiatry (1989) ; 42: : 1267-1272.
[8]	Abdollahi I, Taghizadeh A, Shakeri H, Eivazi M, Jaberzadeh S. The relationship between isokinetic muscle strength and spasticity in the lower limbs of stroke patients. J Bodyw Mov Ther (2015) ; 19: : 284-290.
[9]	An C-M, Jo S-O. Effects of talocrural mobilization with movement on ankle strength, mobility, and weight-bearing ability in hemiplegic patients with chronic stroke: A randomized controlled trial. J Stroke Cerebrovasc Dis (2017) ; 26: : 169-176.
[10]	Andersen LL, Zeeman P, Jørgensen JR, Bech-Pedersen DT, Sørensen J, Kjær M, et al. Effects of intensive physical rehabilitation on neuromuscular adaptations in adults with poststroke hemiparesis. (2011) ; 25: : 2808-2817.
[11]	Avila MA, Romaguera F, Oliveira AB, Camargo PR, Salvini TF. Bilateral impairments of shoulder abduction in chronic hemiparesis: Electromyographic patterns and isokinetic muscle performance. J EMG Kinesiol (2013) ; 23: : 712-720.
[12]	Barbic S, Brouwer B. Test position and hip strength in healthy adults and people with stroke. Arch Phys Med Rehabil (2008) ; 89: : 784-787.
[13]	Bohannon RW. Knee extension power, velocity and torque: Relative deficits and relation to walking performance in stroke patients. Clin Rehabil (1992) ; 6: : 125-131.
[14]	Bohannon RW. Relative decreases in knee extension torque with increased knee extension velocities in stroke patients with hemiparesis. Phys Ther (1987) ; 67: : 1218-1220.
[15]	Brogårdh C, Flansbjer U-B, Lexell J. No specific effect of whole-body vibration training in chronic stroke: A double blind randomized controlled study. Arch Phys Med Rehabil (2012) ; 93: : 253-258.
[16]	Calmels P, Degache F, Courbon A, Roche F, Ramas J, Fayolle-Minon I, et al. The feasibility and the effects of cycloergometer interval-training on aerobic capacity and walking performance after stroke. Preliminary study. Ann Phys Rehab Med (2011) ; 54: : 3-15.
[17]	Carvalho C, Sunnerhagen KS, Willén C. Walking performance and muscle strength in the later stage poststroke: A nonlinear relationship. Arch Phys Med Rehabil (2013) ; 94: : 845-850.
[18]	Chen C-L, Chang K-J, Wu P-Y, Chi C-H, Chang S-T, Cheng Y-Y. Comparison of the effects between isokinetic and isotonic strength training in subacute stroke patients. J Stroke Cerebrovasc Dis (2015) ; 24: : 1317-1323.
[19]	Clark DJ, Condliffe EG, Patten C. Reliability of concentric and eccentric torque during isokinetic knee extension in post-stroke hemiparesis. Clin Biomech (2006) ; 21: : 395-404.
[20]	Cohen JW, Ivanova TD, Brouwer B, Miller KJ, Bryant D, Garland SJ. Do performance measures of strength, balance and mobility predict quality of life and community reintegration after stroke? Arch Phys Med Rehabil (2018) ; 99: : 713-719.
[21]	Coroian F, Jourdan C, Bakhti K, Palayer C, Jaussent A, Picot M-C, et al. Upper limb isokinetic strengthening versus passive mobilization in patients with chronic stroke: A randomized controlled trial. Arch Phys Med Rehabil (2018) ; 99: : 321-328.
[22]	Dehkordi SN, Talebian S, Olyaei G, Montazeri A. Reliability of isokinetic normalized peak torque assessments for knee muscles in post-stroke hemiparesis. Gait Posture (2008) ; 27: : 715-718.
[23]	Dias C, Freire B, Goulart NBA, Onzi ES, Becker J, et al. Muscle architecture and torque production in stroke survivors: An observational study. Top Stroke Rehabil (2017) ; 24: : 206-213.
[24]	Ekiz T, Aslan MD, Özgirgin N. Effects of Kinesio Tape application to quadriceps muscles on isokinetic muscle strength, gait, and functional parameters in patients with stroke. J Rehabil Res Develop (2015) ; 52: : 323-332.
[25]	Ekstrand E, Lexell J, Brogårdh C. Grip strength is a representative measure of muscle weakness in the upper extremity after stroke. Top Stroke Rehabil (2016) ; 23: : 400-405.
[26]	Ekstrand E, Lexell J, Brogårdh C. Isometric and isokinetic muscle strength n the upper extremity can be reliably measured in persons with chronic stroke. J Rehabil Med (2015) ; 47: : 706-713.
[27]	Eng JJ, Kim CM, MacIntyre DL. Reliability of lower extremity strength measures in persons with chronic stroke. Arch Phys Med Rehabil (2002) ; 83: : 322-328.
[28]	Eng JJ, Lomaglio ML, Macinttre DL. Muscle torque preservation and physical activity in individuals with stroke. Med Sci Spors Exerc (2009) ; 41: : 1353-1360.
[29]	Engardt M, Knutsson E, Jonsson M, Sternhag M. Dynamic muscle strength training in stroke patients: Effects on knee extension torque, electromyographic activity, and motor function. Arch Phys Med Rehabil (1995) ; 76: : 419-425.
[30]	Fernandez-Gonzalo R, Nissemark C, Åslund B, Tesch PA, Sojka P. Chronic stoke patients show early and robust improvements in muscle and functional performance to eccentric-overload flywheel resistance training: A pilot study. J Neurongineer Rehabil (2014) ; 11: : 150.
[31]	Flansbjer U-B, Downham D, Lexell J. Knee muscle strength, gait performance, and perceived participation after stroke. Arch Phys Med Rehabil (2006) ; 87: : 974-980.
[32]	Flansbjer U-B, Holmbäck AM, Downham D, Lexell J. What change in isokinetic knee muscle strength can be detected in men and women with hemiparesis after stroke. Clin Rehabil (2005) ; 19: : 514-522.
[33]	Flansbjer U-B, Lexell J, Brogårdh C. Long-term benefits of progressive resistance training in chronic stroke: A 4-year follow-up. J Rehabil Med (2012) ; 44: : 218-221.
[34]	Flansbjer U-B, Miller M, Downham D, Lexell J. Progressive resistance training after stroke: Effects on muscle strength, muscle tone, gait performance and perceived participation. J Rehabil Med (2008) ; 40: : 42-48.
[35]	Freire B, Dias CP, Goulart NBA, de Castro CD, Becker J, Gomes I, et al. Achilles tendon morphology, plantar flexors torque and passive ankle stiffness in spastic hemiparetic stroke survivors. Clin Biomech (2017) ; 41: : 72-76.
[36]	Gray VL, Yang C-L, Waller SM, Rogers MW. Lateral perturbation-induced stepping: Strategies and predictors in persons poststroke. J Neurol Phys Ther (2017) ; 41: : 222-228.
[37]	Hameau S, Bensmail D, Robertson J, Boudarham J, Roche N, Zory R. Isokinetic assessment of the effects of botulinum toxin injection on spasticity and voluntary strength in patients with spastic hemiparesis. Eur J Phys Rehabil Med (2014) ; 50: : 515-523.
[38]	Hamrin E, Eklund G, Hillgreen A-K, Borges O, Hall J, Hellström O. Muscle strength and balance in post-stroke patients. Upsala J Med Sci (1982) ; 87: : 11-26.
[39]	Hsu A-L, Tang P-F, Jan M-H. Analysis of impairments influencing gait velocity and asymmetry of hemiplegic patients after mild to moderate stroke. Arch Phys Med Rehabil (2003) ; 84: : 1185-1193.
[40]	Hsu A-L, Tang P-F, Jan M-H. Test-retest reliability of isokinetic muscle strength of the lower extremities in patients with stroke. Arch Phys Med Rehabil (2002) ; 83: : 1130-1137.
[41]	Hyun CW, Han EY, Im SH, Choi JC, Kim BR, Yoon HM, et al. Hemiparetic knee extension strength and balance function are predictors of ambulatory function in subacute stroke patients. Ann Rehabil Med (2015) ; 39: : 577-585.
[42]	Karatas M, Çetin N, Bayramoglu M, Dilek A. Trunk muscle strength in relation to balance and functional disability in unihemispheric stroke patients. Am J Phys Med Rehabil (2004) ; 83: : 81-87.
[43]	Kim CM, Eng JJ. The relationship of lower-extremity muscle torque to locomotor performance in people with stroke. Phys Ther (2003) ; 83: : 49-57.
[44]	Kim CM, Eng JJ, MacIntyre DL, Dawson J. Effects of isokinetic strength training on walking in persons with stroke: A double-blind controlled pilot study. J Stroke Cerebrvasc Dis (2001) ; 10: : 265-273.
[45]	Kim CM, Kothari DH, Lum PS, Patten C. Reliability of dynamic muscle performance in the hemiparetic upper limb. J Neurol Phys Ther (2005) ; 29: : 9-17.
[46]	Kim S-J, Cho H-Y, Kim K-H, Lee S-M. Effects of ankle biofeedback training on strength, balance, and gait in patients with stroke. J Phys Ther Sci (2016) ; 28: : 2596-2600.
[47]	Knoor S, Brouwer B, Garland J. Validity of the community balance and mobility scale in community-dwelling persons after stroke. Arch Phys Med Rehabil (2010) ; 91: : 890-896.
[48]	Kwong PWH, Ng SSM, Chung RCK, Ng GYF. A structural equation model of the relationship between muscle strength, balance performance, walking endurance and community integration in stroke survivors. Plos One (2017) ; 12: (10): e0185807.
[49]	Lau RWK, Yip SP, Pang MYC. Whole-body vibration has no effect on neuromotor function and falls in chronic stroke. Med Sci Sports Exerc (2012) ; 44: : 1409-1418.
[50]	Lee D-G, Jeong S-K, Kim Y-D. Effects of underwater treadmill walking training on peak torque of the knee in hemiplegic patients. J Phys Ther Sci (2015) ; 27: : 2871-2873.
[51]	Lee S-B, Kang K-Y. The effects of isokinetic eccentric resistance exercise for the hip joint on function in subacute stroke patients. J Phys Ther Sci (2013) ; 25: : 1177-1179.
[52]	Lee SY, Im SH, Kim BR, Han EY. The effects of a motorized aquatic treadmill exercise program on muscle strength, cardiorespiratory fitness, and clinical function in subacute stroke patients. Am J Phys Med Rehabil (2018) ; 97: : 533-540.
[53]	Liao LR, Ng GY, Jones AY, Huang MZ, Pang MY. Whole-body vibration intensities in chronic stroke: A randomized controlled trial. Med Sci Sports Exerc (2016) ; 48: : 1227-1238.
[54]	Lindmark B, Hamrin E. Relation between gait speed, knee muscle torque and motor scores in post-stroke patients. Scand J Caring Sci (1995) ; 9: : 195-202.
[55]	Lindström B, Gerdle B, Forsgren L. Repeated maximum reciprocal knee movements in patients with minimal overt symptoms after ischemic stroke: An evaluation of mechanical performance and emg. Scand J Rehab Med (1998) ; 30: : 47-54.
[56]	Lomaglio MJ, Eng JJ. Muscle strength and weight-bearing symmetry relate to sit-to-stand performance in individuals with stroke. Gait Posture (2005) ; 22: : 126-131.
[57]	Lum PS, Patten C, Kothari D, Yap R. Effects of velocity on maximal torque production in poststoke hemiparesis. Muscle Nerve (2004) ; 30: : 732-742.
[58]	MacIntyre NJ, Rombough R, Brouwer B. Relationships between calf muscle density and muscle strength, mobiity and bone status in stroke survivors with subacute and chronic lower limb hemiparesis. J Musculoskelet Neuronal Interact (2010) ; 10: : 249-255.
[59]	Maeshima S, Ueyoshi A, Osawa A, Ishida K, Kunimoto K, Shimamoto T, et al. Mobility and muscle strength contralateral to hemiplegia from stroke. Am J Phys Med Rehabil (2003) ; 82: : 456-462.
[60]	Marigold DS, Eng JJ, Tokuno CD, Donnelly CA. Contribution of muscle strength and integration of afferent input to postural instability in persons with stroke. Neurorehabil Neural Repaur (2004) ; 18: : 222-229.
[61]	Marque P, Felez A, Puel M, Demonet JF, Guiraud-Chaud-Chaumeil B, Roques CF, et al. Impairment and recovery of left motor function in patients with right hemiplegia. J Neurol Neurosurg Psychiatry (1997) ; 62: : 77-81.
[62]	Nakamura R, Hosokawa T, Tsuji I. Relationship of muscle strength for knee extension to walking capacity in patients with spastic hemiparesis. Tohoku J Exp Med (1985) ; 145: : 335-340.
[63]	Nascimento LR, Teixeira-Salmela LF, Polese JC, Ada L, Faria CDCM, Laurentino GEC. Strength deficits of the shoulder complex during isokinetic testing in people with chronic stroke. Braz J Phys Ther (2014) ; 18: : 268-275.
[64]	Novak AC, Brouwer B. Strength and aerobic requirements dusing stair ambulation in persons with chronic stroke and healthy adults. Arch Phys Med Rehabil (2012) ; 93: : 683-689.
[65]	Pang MYC, Lau RWK, Yip SP. The effects of whole-body vibration therapy on bone turnover, muscle strength, motor function, and spasticity in chronic stroke: A randomized controlled trial. Eur J Phys Rehabil Med (2013) ; 49: : 439-450.
[66]	Ploutz-Snyder LL, Clark BC, Logan L, Turk M. Evaluation of spastic muscle in stroke survivors using magnetic resonance imaging and resistance to passive motion. Arch Phys Med Rehabil (2006) ; 87: : 1636-1652.
[67]	Pohl PS, Startzell JK, Duncan PW, Wallace D. Reliability of lower extremity strength testing adults with stroke. Clin Rehabil (2000) ; 14: : 601-607.
[68]	Quintino LF, Franco J, Gusmão AFM, De Souza Silva PF, Morais Faria CDC. Trunk flexor and extensor muscle performance in chronic stroke patients: A case-control study. Braz J Phys Ther (2018) ; 22: : 231-237.
[69]	Rabelo M, Fachin-Martins E. Inter-rater and test/retest reliabilities of the isokinetic measurements: Assessing strength and endurance of the trunk muscles in two different protocols for able-bodies and post-stroke hemiparesis. Top Stroke Rehabil (2018) ; 25: : 424-431.
[70]	Ryan AS, Buscemi A, Forrester L, Hafer-Macko CE, Ivey FM. Atrophy and intramuscular fat in specific muscles of the thigh: Associated weakness and hyperinsulinemia in stroke survivors. Neurorehabil Neural Repair (2011) ; 25: : 865-872.
[71]	Şen SB, Özbudak DS, Demir SÖ, Ekiz T, Özgirgin N. Effects of the bilateral isokinetic strengthening training on functional parameters, gait, and the quality of life in patients with stroke. Int J Clin Exp Med (2015) ; 8: : 16871-16879.
[72]	Sharp SA, Brouwer BJ. Isokinetic strength training of the hemiparetic knee: Effects on function and spasticity. Arch Phys Med Rehabil (1997) ; 78: : 1231-1236.
[73]	Shimodozono M, Kawahira K, Ogata A, Etoh S, Tanaka N. Addition of a anabolic steroid to strength training promotes muscle strength in nonparetic lower limb of poststroke hemiplegia patients. Int J Neurosci (2010) ; 120: : 617-624.
[74]	Silva P, Franco J, Gusmão A, Moura J, Teixeira-Salmela L, Faria C. Trunk strength is associated with sit-to-stand performance in both stroke and healthy subjects. Eur J Phys Rehabil Med (2015) ; 15: : 717-724.
[75]	Sunnerhagen KS, Mattsson K. One-legged bicycling as an assessment tool for patients with stroke. Acta Neurol Scand (2005) ; 111: : 373-378.
[76]	Suzuki K, Imada G, Iwaya T, Handa T, Kurogo H. Determinants and predictors of the maximum walking speed dusing computer-assisted gait training in hemiparetic stroke patients. Arch Phys Med Rehabil (1999) ; 80: : 179-182.
[77]	Suzuki K, Nakamura R, Yamada Y, Handa T. Determinants of maximum walking speed in hemiparetic stroke patients. Tohoku J Exp Med (1990) ; 163: : 337-344.
[78]	Tanaka S, Hachisuka H, Ogata H. Muscle strength of trunk flexion-extension in post-stroke hemiplegic patients. Am J Phys Med Rehabil (1998) ; 77: : 288-290.
[79]	Tanaka S, Kenji H, Ogata H. Trunk rotatory muscle performance in post-stroke hemiplegic patients. Am J Phys Med Rehabil (1997) ; 76: : 366-369.
[80]	Tankisheva E, Bogaerts A, Boonen S, Feys H, Verschueren S. Effects of intensive whole-body vibration training on muscle strength and balance in adults with chronic stroke: A randomized controlled pilot study. Arch Phys Med Rehabil (2014) ; 95: : 439-446.
[81]	Teixeira-Salmela LF, Olney SJ, Nadeau S, Brouer B. Muscle strengthening and physical conditioning to reduce impairment and disability in chronic stroke survivors. Arch Phys Med Rehabil (1999) ; 80: ; 1211-1218.
[82]	Tyson SF, Sadeghi-Demneh E, Nester CJ. The effects of transcutaneous electrical nerve stimulation on strength, proprioception, balance and mobility in people with stroke: A randomized controlled cross-over trial. Clin Rehabil (2013) ; 27: : 785-791.
[83]	Wang W-T, Huang L-T, Chou Y-H, Wei T-S, Lin C-C. Nonparetic knee extensor strength is the determinant of exercise capacity of community-dwelling stroke survivors. Scientific World 2014, Article ID 769875.