Correlation of paretic knee extension strength with gait speed after stroke: A meta-analysis

Abstract

BACKGROUND AND OBJECTIVE:

Numerous studies have addressed the relationship between paretic knee extension strength and comfortable walking speed after stroke. However, the correlations reported are highly variable. This review sought to summarize the correlational data using meta-analysis.

METHODS:

Relevant literature was identified via a search of 3 bibliographic databases. Articles were screened and perused for inclusion. Included articles were examined for information on the sample studied, procedures for measuring strength and gait speed, and correlations reported between the 2 variables. Meta-analysis was used to calculate a summary correlation.

RESULTS:

Of 299 unique articles, 18 met inclusion criteria. Articles were diverse in regard to samples studied, procedures described, and correlations reported. Meta-analysis using data from all included studies revealed a summary correlation of 0.51. For studies using hand-held or isokinetic dynamometry, the summary correlation was 0.46 and 0.59 respectively.

CONCLUSIONS:

This study provides a better indication of the correlation between paretic knee extension strength and comfortable gait speed than individual studies. The correlation is high enough to provide support for the routine measurement of paretic knee extension force for individuals who have experienced a stroke.

Keywords

Muscle strength gait

1. Introduction

Reduced muscle strength is the preeminent impairment in most patients who have experienced a stroke [1]. The impairment is most obvious in, but not limited to, the limb muscles of the paretic side [2]. Research addressing muscle strength after stroke often addresses knee extension. This focus is logical given the relative simplicity of measuring knee extension strength and its relationship with performance of activities such as standing from sitting [3] maintaining standing [4] transferring from bed/mat-to-chair [5], and walking [6]. Research examining the relationship between knee extension strength and comfortable walking speed is particularly plentiful [7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24]. As research on this relationship is diverse in regard to procedures and outcomes, this systematic review was conducted to consolidate findings relative to the relationship. Hopefully my findings will provide guidance to clinicians working with patients with lower limb paresis and gait limitations after stroke.

2. Methods

This review involved electronic searches of the PubMed, Scopus, and CINAHL databases by the author and is valid for February 28, 2021. The searches employed the search string “stroke AND strength AND knee AND (gait OR walk).” A manual search of reference lists and files followed. To be eligible for inclusion in the review an article had to specifically address the correlation between objective measures of paretic knee extension strength (force or moment) and comfortable gait speed in patients with stroke. Articles dealing with other aspects of muscle performance (e.g., power) and gait speed (e.g., maximum) instead did not qualify an article for inclusion. Reviews, case studies, dissertations, and articles in languages other than English were excluded.

Articles qualifying for inclusion were perused for information on the sample tested, specifics of how strength and gait speed were measured, and correlations reported between knee extension strength and gait speed. In cases in which more than one correlation was reported, the first was used in subsequent meta-analysis.

A meta-analysis of correlations was conducted using MedCalc statistical software [25]. Specifically, a summary correlation coefficient was calculated using a random effects model that incorporated the sample sizes and correlation coefficients from individual studies.

3. Results

Of 299 nonduplicative articles considered, 18 were eventually found to meet all criteria for inclusion and none of the criteria for exclusion (Fig. 1). These studies were conducted primarily in the United States, but they originated in Brazil and 5 other countries as well. Tested individuals resided in a variety of settings-from acute rehabilitation to community-dwelling. The total number of participants was 624 with individual studies contributing correlations from 12 to 90 participants (Table 1).

Table 1
Cross-sectional correlations between knee extension strength and gait speed of patients with stroke

Reference	Sample	Strength measurement	Speed measurement	Correlation
Aguiar et al. [7]	Brazilian Community-dwelling, ( $n=$ 41)	Gravity-eliminated isometric force measured with HHD.	10.0 m with dynamic start @ “comfortable” speed.	Single session $=$ 0.40
Bohannon [8]	American Rehabilitation inpatients ( $n=$ 20)	Gravity-eliminated isometric force measured with HHD and normalized against body weight.	8.0 m with dynamic start@ “most comfortable speed.”	Single session $=$ 0.36
Bohannon [9]	American Rehabilitation inpatients ( $n=$ 33)	Gravityeliminated isometric force measured with HHD and normalized against body weight.	8.0 m with dynamic start @ “comfortable pace.”	Initial session $=$ 0.81; final session $=$ 0.73
Bohannon [10]	American Rehabilitation inpatients ( $n=$ 20)	Gravity eliminated isometric force measured with HHD & normalized against body weight.	10.0 m with dynamic start @ “most comfortable speed”	Single session $=$ 0.67
Bohannon [11]	American Rehabilitation inpatients and outpatients ( $n=$ 20)	Gravity eliminated isometric torque measured with IKD	7.0 m with dynamic start @ “most comfortable speed”	Single session $=$ 0.75
Bohannon and Andrews [12]	American Community-dwelling ( $n=$ 17)	Gravity eliminated isometric torque measured with IKD	8.0 m @ “comfortable pace.”	First day $=$ 0.61 & 0.54; second day $=$ 0.56 & 0.58
Dorsch et al. [13]	Australian Community-dwelling ( $n=$ 60)	Isometric force measured with HHD	10.0 m with dynamic start @ “comfortable” speed	Single session $=$ 0.27
Flansbjer et al. [14]	Swedish Community-dwelling ( $n=$ 50)	Torque measured concentrically @ 60 ${}^{\circ}$ /s with IKD	10.0 m with dynamic start @ “self-selected comfortable pace ‘like walking in a park.”’	Single session $=$ 0.61
Hsu et al. [15]	Taiwanese Rehabilitation outpatients ( $n=$ 26)	Torque measured concentrically @ 30 ${}^{\circ}$ /s with IKD.	3.8 m with dynamic start at “comfortable” speed.	Single session $=$ 0.52
Hyun et al. [16]	Korean Subacute patients ( $n=$ 53)	Gravity corrected torque measured isometrically @ 60 ${}^{\circ}$ with IKD	10.0 m @ “comfortable but fastest pace safely possible.”	Single session $=$ 0.55
Isho and Usuda [17]	Japanese Acute rehabilitation inpatients ( $n=$ 15)	Gravity eliminated isometric force measured with HHD	10.0 m with dynamic start @ “preferred comfortable speed”	Single session $=$ 0.26
Kim and Eng [18]	Korean Chronic patients ( $n=$ 20)	Torque measured concentrically @ 60 ${}^{\circ}$ /s with IKD.	8.0 m @ “most comfortable speed”	Single session $=$ 0.41
LeBrasseur et al. [19]	American Community-dwelling ( $n=$ 31)	One repetition maximum measured with pneumatic resistance machine.	10.0 m @ “habitual” speed	Single session $=$ 0.36
Lindmark and Hamrin [20]	Swedish Patients 3 months post-discharge ( $n=$ 34)	Torque measured @ 90 ${}^{\circ}$ /s with IKD.	5.0 m with dynamic start @ “freely chosen speed.”	Single session $=$ 0.53 (men) and 0.21 (women)
Menezes et al. [21]	Brazilian Community-dwelling ( $n=$ 90)	Isometric force measured with HHD.	10. m dynamic start @ “most comfortable speed.”	Single session $=$ 0.32
Nasciutti-Prudente et al. [22]	Brazilian Outpatient rehabilitation ( $n=$ 12)	Gravity-corrected torque measured with HHD	10.0 m @ “self-selected speed.”	Single section $=$ 0.34
Patterson et al. [23]	American Community-dwelling ( $n=$ 74)	Torque measured eccentrically @ 30 ${}^{\circ}$ /s, 90 ${}^{\circ}$ /s & 120 ${}^{\circ}$ /s with IKD to derive a composite score.	30 ft @ “self-selected comfortable” speed.	Single session $=$ 0.60
Wang et al. [24]	Taiwanese Chronic patients( $n=$ 35)	Torque measured concentrically @ 60 ${}^{\circ}$ /s & 90 ${}^{\circ}$ /s with IKD	12 m @ “comfortable pace”	Single session $=$ 0.62 @ 60 ${}^{\circ}$ s & 0.66 @ 90 ${}^{\circ}$ /s.

${}^{*}$ HHD $=$ hand-held dynamometer, IKD $=$ isokinetic dynamometer. Correlations used in meta-analysis are underlined.

Figure 1.

Study flow chart.

With the exception of one study, knee extension strength was measured using hand-held dynamometry or isokinetic dynamometry. Measurements were reported in units of force or torque with some adjusted for gravity or participant weight. Gait speed was typically measured after patients were instructed to walk at a “comfortable” speed, though synonymous terms (e.g. “habitual”) were sometimes used. Most gait speed tests involved a dynamic start. The timed distances ranged from 3.8 m to 12 m. (Table 1).

The correlations between paretic knee extension strength and gait speed ranged from 0.21 to 0.81 (Table 1). The meta-analysis determined a summary correlation of 0.51 (95% confidence intervals $=$ 0.43 to 0.59 (Fig. 2). Heterogeneity as indicated by $I^{2}$ (42.4%), was acceptable. Egers test ( $p=$ 0.63) and a funnel plot (Fig. 3) did not indicate sample size to be a source of bias. The summary correlation was 0.46 for studies employing hand-held dynamometry and 0.59 for studies employing isokinetic dynamometry. The 2 correlations were significantly different ( $z=$ $-$ 2.16, $p=$ 0.031).

Figure 2.

Forrest plot of studies included in review.

Figure 3.

Funnel plot of studies included in review.

4. Discussion

The objective of this review was to consolidate data from multiple studies to establish the relationship between paretic knee extension strength and comfortable gait speed in patients with stroke. The summary correlation calculated by meta-analysis (0.51) should provide a better indication of the relationship between the variables than the individual studies included. The moderate correlation calculated using meta-analysis is of sufficient magnitude to support recommendation of the objective measurement of paretic knee extension after stroke. This support is present regardless of whether strength measurements were obtained by hand-held and isokinetic dynamometry. That noted, measures of knee extension strength obtained by isokinetic dynamometry may provide a better explanation of gait speed than those obtained using hand-held dynamometry. This could be because of inadequate tester strength [26].

The value of the findings of this meta-analysis notwithstanding, they are narrow in scope. The strength of lower limb muscle actions other than paretic knee extension (including those from the nonparetic side) [9, 24]; and combinations of muscle action strengths [27] have also been found to correlate with comfortable gait speed. Moreover, other measures of muscle performance such as power have been shown to correlate with comfortable gait speed [11]. Finally, muscle strength is known to relate with maximum gait speed also [6].

This meta-analysis had several procedural limitations. Among these were use of a single individual to search for and vett articles for inclusion. No quality rating was performed other than making certain that inclusion and exclusion criteria were addressed.

5. Conclusions

Numerous studies have examined the relationship between paretic knee extension strength and comfortable walking speed. This meta-analysis of the studies shows that the correlation between the variables is moderate. Given the numerous factors contributing to successful ambulation, the correlation is of sufficient magnitude to warrant the routine objective measurement and treatment of paretic knee extension strength after stroke.

Ethical considerations

As a review, this study is exempt from IRB approval.

Funding

None.

Footnotes

Acknowledgments

None.

Conflict of interest

None. Given his role as an Editorial Board Member, Richard W. Bohannon had no involvement nor access to information regarding the peer review of this article.

References

Lawrence

Coshall

Dundas

Stewart

Rudd

Howard

Wolf

. Estimates of prevalence of acute stroke impairments and disability in a multiethnic population. Stroke 2001; 32(6): 1279-1284.

Bohannon

Andrews

. Limb muscle strength is impaired bilaterally after stroke. J Phys Ther Sci 1995; 7(1): 1-7.

Bohannon

. Knee extension strength and body weight determine sit-to-stand independence after stroke. Physiother Theory Pract 2007; 23(5): 291-297.

Bohannon

. Correlation of lower limb strengths and other variables with standing performance in stroke patients. Physiother Can 1989; 41(4): 198-202.

Bohannon

. Determinants of transfer capacity in patients with hemiparesis. Physiother Can 1988; 40(4): 236-239.

Nakamura

Hosokawa

Tsuji

. Relationship of muscle strength for knee extension to walking capacity in patients with spastic hemiparesis. Tohoku J Exp Med 1985; 145(3): 335-340.

Aguiar

Camargo

LBS

Estarlino

Texeira-Salmela

de Morais Faria

CDC

. Strength of lower limb and trunk muscles is associated with gait speed in individuals with sub-acute stroke: a cross-sectional study. Braz J Phys Ther 2018; 22(6): 459-477.

Bohannon

. Strength of lower limb related to gait velocity and cadence in stroke patients. Physiother Can 1986; 38(4): 204-206.

Bohannon

. Selected determinants of ambulatory capacity in patients with hemiplegia. Clin Rehabil 1989; 3(1): 47-53.

10.

Bohannon

. Relationship among paretic knee extension strength, maximum weight-bearing, and gait speed in patients with stroke. J Stroke Cerebrovasc Dis 1991; 1(2): 65-69.

11.

Bohannon

. Knee extension power, velocity and torque: relative deficits and relation to walking performance in stroke patients. Clin Rehabil 1992; 6(2): 125-131.

12.

Bohannon

Andrews

. Correlation of knee extensor muscle torque and spasticity with gait speed in patients with stroke. Arch Phys Med Rehabil 1990; 71(5): 330-333.

13.

Dorsch

Ada

Canning

Al-Zharani

Dean

. The strength of the ankle dorsiflexors has a significant contribution to walking speed in people who can walk independently after Stroke: An observational study. Arch Phys Med Rehabil 2012; 93(6): 1072-1076.

14.

Flansbjer

U-B

Downham

Lexell

. Knee muscle strength, gait performance, and perceived participation after stroke. Arch Phys Med Rehabil 2006; 87(7): 974-980.

15.

Hsu

A-L

Tang

P-F

Jan

M-H

. Analysis of impairments influencing gait velocity and asymmetry of hemiplegic patients after mild stroke. Arch Phys Med Rehabil 2003; 84(8): 1185-1193.

16.

Hyun

Han

Choi

Kim

Yoon

, et al. Hemiparetic knee extensor strength and balance function are predictors of ambulatory function in subacute stroke patients. Ann Rehabil Med 2015; 39(4): 577-585.

17.

Isho

Usuda

. Association of trunk control with mobility performance and accelerometry-based gait characteristics in hemiparetic patients with subacute stroke. Gait Posture 2016; 44(2): 89-93.

18.

Kim

Eng

. The relationship of lower-extremity muscle torque to locomotor performance in people with stroke. Phys Ther 2003; 83(1): 49-57.

19.

LeBrasseur

Sayers

Ouellette Fielding

. Muscle impairments and behavioral factors mediate functional limitations and disability following stroke. Phys Ther 2006; 86(10): 1342-1350.

20.

Lindmark

Hamrin

. Relation between gait speed, knee muscle torque and motor scores in post-stroke patients. Scand J Caring Sci 1995; 9(4): 195-202.

21.

Menezes

KKP

Nascimento

Faria

CDCM

Avelino

Scianni

, et al. Deficits in motor coordination of the paretic lower limb best explained activity limitations after stroke. Physiother Theory Pract 2020; 36(3): 417-423.

22.

Nasciutti-Prudente

Oliveira

Houri

de Paula Goulart

Neto

Teixeira-Salmela

. Relationship between muscular torque and gait speed in chronic hemiparetic subjects. Disabiul Rehabil 2009; 31(2): 103-108.

23.

Patterson

Forrester

Rodgers

Ryan

Ivey

Sorkin

, et al., Determinants of walking function after stroke: differences by deficit severity. Arch Phys Med Rehabil 2007; 88(1): 115-119.

24.

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 J 2014; ID 769875.

25.

Schoonjans

. Medcalc Manual. Ostend, Belgium, Medcalc Software, 2017.

26.

Wikholm

Bohannon

. Hand-held dynamometer measurements: tester strength makes a difference. J Orthop Sports Phys Ther 1991; 13(4): 191-198.

27.

Bohannon

. Gait performance of hemiparetic stroke patients: selected variables. Arch Phys Med Rehabil 1987; 68(11): 777-781.

Correlation of paretic knee extension strength with gait speed after stroke: A meta-analysis

Abstract

BACKGROUND AND OBJECTIVE:

METHODS:

RESULTS:

CONCLUSIONS:

Keywords

1. Introduction

2. Methods

3. Results

Table 1 Cross-sectional correlations between knee extension strength and gait speed of patients with stroke

5. Conclusions

Ethical considerations

Funding

Footnotes

Acknowledgments

Conflict of interest

References

Table 1
Cross-sectional correlations between knee extension strength and gait speed of patients with stroke