Abstract
Purpose
-
The aim of the study was to quantify motor cortex descending drive and voluntary activation (VA) in people with lower-limb OA compared to controls.
Methods
-
A systematic review and meta-analysis according to the PRISMA guidelines was carried out. Seven databases were searched until 30 December 2022. Studies assessing VA or responses to transcranial magnetic stimulation (TMS; i.e. motor evoked potential, intracortical facilitation, motor threshold, short-interval intracortical inhibition, and silent period) were included. Study quality was assessed using Joanna Briggs Institute criteria and evidence certainty using GRADE. The meta-analysis was performed using RevMan inverse variance, mixed-effect models.
Results
-
Eighteen studies were included, all deemed low-quality. Quadriceps VA was impaired with knee OA compared to healthy controls (standardised mean difference (SMD) = 0.84, 95% CI = −1.12–0.56, low certainty). VA of the more symptomatic limb was impaired (SMD = 0.42, 95% CI = −0.75–0.09, moderate certainty) compared to the other limb in people with hip/knee OA. As only two studies assessed responses to TMS, very low-certainty evidence demonstrated no significant difference between knee OA and healthy controls for motor evoked potential, intracortical facilitation, resting motor threshold or short-interval intracortical inhibition.
Conclusions
-
Low-certainty evidence suggests people with knee OA have substantial impairments in VA of their quadriceps muscle when compared to healthy controls. With moderate certainty we conclude that people with hip and knee OA had larger impairments in VA of the quadriceps in their more painful limb compared to their non-affected/other limb.
Introduction
Osteoarthritis (OA) is a leading cause of pain and disability globally (1). It results in substantial health expenditure in Australia, with over $3.5 billion being spent annually on OA burden (2). OA presents as pain and impaired function, causing people to leave the workforce early and frequently progress to pharmacological and surgical management (3, 4). Best practice management for OA includes exercise targeted at peripheral impairments (e.g. muscle strength, range of motion, and proprioception) (5). However, the pain experienced with OA is complex and not solely a result of nociceptive input from peripheral tissue (6). By understanding different central nervous system contributions to motor dysfunction in OA, we can provide more targeted interventions to improve pain, muscle weakness and function.
A major misconception with OA is that pain severity and function are related to the degree of structural damage (7). Evidence actually shows that pain severity with OA has a poor relationship with the degree of structural damage visible on imaging (8). Similar to other musculoskeletal conditions (9), improvements in pain experienced can occur whilst visible ‘damage’ remains unchanged (10). It has been proposed that pain improvements occur due to improved neuromuscular function with exercise rehabilitation (11). However, this hypothesis has been challenged in other musculoskeletal conditions, with changes in disability being unrelated to neuromuscular function (12). Specifically in OA, improvements in muscle function, including strength, are proposed to improve the joint’s capacity to absorb impact forces and thereby reduce joint load (10).
When attempting to actively contract their muscles, people with persistent pain (i.e. OA) have impaired voluntary activation (13, 14). However, we do not know the mechanism by which this impaired activation occurs. It may be that altered descending neural drive and subsequently greater inhibition contributes to impaired voluntary maximal force production in people with OA (15). This may then reduce exercise performance or reduce capacity to increase muscular strength (16), the primary aim of exercise rehabilitation. Conversely, a large meta-analysis of chronic pain conditions demonstrated disinhibition, with subsequent increases in motor cortex excitability (17). The function of the motor cortex, responsible for initiating and providing descending drive to allow voluntary muscle contraction, is typically assessed using transcranial magnetic stimulation (TMS), which provides insight into the level of motor cortex excitability and inhibition.
In knee OA, one study (n = 107) demonstrated that decreased intracortical facilitation (ICF) and a reduced motor threshold (MT) were associated with increased pain when assessed using a visual analogue scale (16). Conversely, this same study reported that when using the Western Ontario and McMaster Universities Arthritis Index (WOMAC) pain scale, increased ICF and increased MT were associated with increased pain (16). These conflicting findings make interpretation of the results challenging. Therefore, a comprehensive understanding of motor cortex drive, including other measures of corticospinal and intracortical excitability, and overall neural activation of the muscle, may improve the understanding of the mechanisms contributing to motor dysfunction in OA. This may assist exercise (and other) intervention selection, dosage, and parameters to improve OA outcomes. The aim of this review was to determine the extent that people with lower-limb OA have altered neural responses implicated in descending motor drive and voluntary muscle activation.
Objectives
Our primary objective was to quantify motor cortex descending drive and voluntary activation in people with lower-limb OA compared to controls.
Methods
Guidelines and prospective registration
This systematic review and meta-analysis was designed and reported as per the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) recommendations (18). The protocol was registered prospectively with PROSPERO (CRD42022381635; https://www.crd.york.ac.uk/prospero/).
Data management
All records were managed and stored within Covidence (Veritas Health Innovation, Melbourne, Australia). Study data extraction utilised Microsoft Excel spreadsheets. To facilitate systematic review transparency (19, 20), all extracted data are included within this submission.
Inclusion criteria
Participants
Our pathological group included humans aged ≥18 years, who had been diagnosed with lower-limb OA (hip, knee, ankle, and/or foot). Studies whose participants had mixed presentations of joint OA were included, provided the region of OA was identified.
For studies utilising TMS or assessing voluntary activation, our control group included reportedly healthy humans aged ≥18 years who did not have lower-limb OA but were not required to have been matched for age, sex, or physical activity levels. For studies assessing voluntary activation we also included studies that assessed the contralateral non/less affected limb as a control, however these were assessed separately to healthy controls.
Outcomes
The primary outcome for this review was the assessment of measures that assess aspects of descending motor drive. We included studies that reported any measure inclusive of cervicomedullary evoked potentials (CMEP), ICF, motor evoked potential (MEP), active motor threshold (AMT), resting motor threshold (RMT), short-interval intracortical inhibition (SICI), and silent period (SP). Furthermore, studies that assessed voluntary activation (the ability of the nervous system to drive the muscle to produce maximal force) were included. A detailed description of each of these terms is included within Appendix A (see section on supplementary materials given at the end of this article).
Types of studies
Cross-sectional, case–control, and observational studies were included. We also included randomised controlled trials (RCTs) that performed analysis at baseline (e.g. some RCTs performed a cross-sectional analysis of voluntary activation at baseline versus healthy controls). Studies were included regardless of their publication status. Reviews and case reports were excluded. We included records in all languages, which were translated as needed.
Search strategy
Search strategies using free text terms (Appendix B) were performed from inception to 30 December 2022. Searches were performed within the following electronic databases: PubMed, CINAHL (Full-text), EBSCO (Medline), Cochrane library, SPORTDiscus, Web of Science (Appendix C). Clinical trial registries were not searched as we examined cross-sectional data only. Reference lists of reviews and retrieved articles were checked for additional studies missed in the electronic database search. The ePublication lists of key journals in the field were screened to identify studies that have yet to be indexed.
Study selection
Identified studies were exported to reference management software, EndNoteTM 20 (Version 20.4.1, Clarivate Analytics, Philadelphia, PA, USA) and then uploaded into Covidence, with duplicates automatically removed by the software. Two review authors (WH and MM) independently assessed the titles and abstracts of eligible studies identified by the search strategy. Articles that met the inclusion criteria were assessed in full. Disagreements were resolved by consensus.
Data extraction
Two review authors (MM and CS) independently extracted data from all included studies written in English. For studies that were translated from another language, a single study author (SM) performed data extraction directly from the manuscript to minimise errors, under the guidance of one study author (MM). Discrepancies and disagreements were resolved by consensus. The following data were extracted: primary author; year of publication; country in which study was conducted; study design; diagnosis; clinical diagnosis was reported; radiological diagnosis was reported; sample size; age; height; weight; body mass index; sex; gender; duration of pain; co-morbidities; injury history; TMS and voluntary activation assessment protocol(s); ICF; MEP; RMT; AMT; SICI; SP; voluntary activation; quality of life; disability; physical function; severity of OA on imaging (Kellgren–Lawrence grading system: 0 = none; 1 = doubtful; 2 = minimal; 3 = moderate; 4 = severe) (21).
Data management
Where presented in graphical format, data were extracted using an online software tool that provided the mean and measure of variance (https://apps.automeris.io/wpd/). Where median (interquartile range: IQR) was presented, it was converted to mean (s.d.) by assuming the median was the same as the mean and s.d. was 1.35 times the IQR (22).
Data synthesis
Demographic data were described using count, percentage, mean, s.d. (or non-parametric alternatives), as indicated. Meta-analyses of between-group outcomes were performed using an inverse variance, random-effects models in RevMan (ReviewManager program, Version 5.4, The Cochrane Collaboration, 2020). For voluntary activation, the pooled standardised mean difference (SMD) with 95% confidence intervals (95% CI) was calculated as the units of measure were not comparable for all studies. For TMS the SMD (95% CI) was calculated as one study used the lower-limb as the reference site and the other study used the upper limb.
Assessment of methodological quality
Two review authors (CL and MM) independently assessed risk of bias for each study using the Joanna Briggs Institute Checklist for Analytical Cross-sectional Studies with any disagreements being resolved by consensus or a third review author if consensus could not be reached. This scale includes eight items, and we defined criteria for each item a priori (Appendix D). An overall judgement of methodological quality was assigned based on a ‘worst-item-counts’ basis with studies being assigned ‘low’ quality if at least one item is reported as ‘no’, unclear if no items are reported as ‘no’ and at least one item is reported as ‘unclear’, and ‘high’ quality if all items are reported as ‘yes’.
Statistical heterogeneity
A chi-square test was used to evaluate statistical heterogeneity (22). The presence of heterogeneity was assessed using the I² statistic, based on the P-value being <0.10 or the I² value being >40% (22).
Consideration of other biases
The influence of small study biases were addressed. Studies with sample sizes less than 50 cases were considered as representing high risk, studies with samples between 50 and 200 cases were classified as moderate risk, and studies with sample sizes greater than 200 cases were classed as low risk (23, 24, 25).
Funnel plots for each variable were visually inspected to explore the likelihood of reporting biases when at least 10 studies were included in a meta-analysis (26).
Assessment of the quality of the body of evidence
The quality of the body of evidence was determined using the Grading of Recommendations, Assessment, Development, and Evaluations (GRADE) approach, which was adapted for this study design. The GRADE approach involves making an overall judgement on the quality of the body of evidence based on the overall risk of bias, consistency of results, precision, directness of the evidence and publication bias.
Deviations to protocol
Analysis using individual participant data had been planned, however due to the age of included studies (most studies published prior to 2005) and only two studies assessing responses to TMS, we did not proceed with individual participant data meta-analysis.
Results
Selection of studies
A total of 1578 records were identified, and after removal of duplicates, 41 (n = 41) records proceeded to full-text review. Finally, 18 (n = 18) records were included in our review (Fig. 1). Reasons for full-text exclusion are reported in Appendix E.
Study information
Full study data are presented in Table 1. All studies reported that their participants were clinically diagnosed with OA (13, 14, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42); however, the criteria for diagnosis varied substantially between studies. 78% of studies also reported that radiological investigations supported the clinical diagnosis (13, 14, 27, 28, 29, 30, 31, 33, 34, 37, 38, 39, 40, 41). Knee OA was investigated in 17/18 studies (94%) (13, 14, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 39, 40, 41, 42), whilst the remaining study assessed hip OA (38). Nine studies (50%) compared findings of the symptomatic/most symptomatic limb to the asymptomatic/less symptomatic limb (27, 28, 34, 35, 36, 37, 38, 41, 42) with 11 studies (61%) comparing the OA group to a healthy control (13, 14, 27, 28, 29, 30, 31, 32, 33, 39, 40). Voluntary activation of the quadriceps was assessed in 17/18 studies (94%) (13, 14, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 40, 41, 42), and two studies assessed responses to TMS (32, 39) (Appendix F for a detailed description of these assessment techniques). The majority of the studies were from the USA (6/18, 33%) (32, 33, 35, 36, 37, 40), followed by those from Germany (4/18, 22%) (13, 14, 27, 34). All but one study (94%) (13, 14, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41) received financial support (Appendix G).
Study information.
| Study | Country | Diagnosis | Co-morbidities | Sample injury history | Limb compared to | |
|---|---|---|---|---|---|---|
| AS/LS limb* | HL of control† | |||||
| Berth et al. (27) | Germany | Unilateral knee OA | NR | NR | Y | Y |
| Gapeyeva et al. (28) | Estonia | Unilateral knee OA | NR | NR | Y | Y |
| Hassan et al. (29) | UK | Bilateral or unilateral knee OA | NR | NR | Y | |
| Heiden et al. (30) | Australia | Knee OA | NR | NR | Y | |
| Hurley et al. (31) | UK | Bilateral or unilateral knee OA | NR | NR | Y | |
| Kittelson et al. (32) | USA | Unilateral knee OA | NR | NR | Y | |
| Lewek et al. (33) | USA | Unilateral knee OA (medial compartment) with genu varum | NR | NR | Y | |
| Machner et al. (34) | Germany | Unilateral knee OA (primarily the medial tibiofemoral joint) | NR | NR | Y | |
| Pap et al. (13) | Germany | Unilateral knee OA with genu varum | NR | NR | Y | |
| Pap et al. (14) | Germany | Unilateral knee OA (medial and lateral femoral and tibial condyle) | NR | NR | Y | |
| Petterson et al. (35) | USA | Unilateral knee OA | NR | NR | Y | |
| Petterson et al. (36) | USA | Unilateral knee OA | NR | NR | Y | |
| Stevens et al. (37) | USA | Unilateral OA (tricompartmental) | NR | NR | Y | |
| Suetta et al. (38) | Denmark | Unilateral hip OA | NR | NR | Y | |
| Tarrago et al. (39) | Brazil | Unilateral knee OA | HT: 47.6%, DB: 9.5%, Asthma: 9.5% | NR | Y | |
| Thomas et al. (40) | USA | Unilateral or bilateral knee OA | NR | NR | Y | |
| Vahtrik et al. (41) | Estonia | Unilateral or bilateral mild knee OA | NR | Heavy physical work or prior trauma | Y | |
| Ventura et al. (42) | Switzerland | Unilateral knee OA (medial tibiofemoral or multicompartmental) | NR | NR | Y | |
*Compared symptomatic/most symptomatic limb to the asymptomatic/less symptomatic limb; †Compared the symptomatic/most symptomatic limb to the healthy limb of a control.
AS, asymptomatic; DB, diabetes; HL, healthy limb; HT, hypertension; LS, less symptomatic; OA, osteoarthritis; NR, not reported.
Participant demographics
Complete study level demographic details are presented in Tables 2 and 3. A total of 1271 participants were included in the articles (913 OA and 358 controls) with sample sizes in OA groups ranging from 10 (28) to 154 (14) participants. Female inclusion between studies varied, ranging between 29% (38) and 100% females (28). Mean OA participant ages ranged from 52.6 (33) to 67.3 (34) years, with mean control ages often being younger (e.g. two control cohorts mean ages were <35 years (32, 39)). Mean BMI also varied for OA participants, ranging from 25.1 (14) to 33.0 (40), with mean control BMI often lower (e.g. three control cohorts BMI were <25 (29, 30, 32)). The severity of OA also varied, with studies including participants ranging from grade II to grade IV OA. Two studies (11%) included quality of life (30, 42) and nine studies (50%) included a measure of disability (14, 28, 29, 30, 31, 37, 38, 40, 42). Only two studies (11%) reported symptom duration (14, 28).
Participant baseline demographics.
| Study/group | Sample size, n | Female (%) | Age*, years | Height*, cm | Weight*, Kg | BMI* |
|---|---|---|---|---|---|---|
| Berth et al. (27) | ||||||
| OA | 50 | 64 | 65.8 (NR) | 31.0 (NR) | ||
| CONT | 46 | 65 | 63.2 (NR) | 26.2 (NR) | ||
| Gapeyeva et al. (28) | ||||||
| OA | 10 | 100 | 63.0 (52.0-74.0) | 156.0 (151.0-166.0) | 73.0 (55.0-101.0) | 29.0 (22.0-38.0) |
| CONT | 10 | 100 | 64.0 (52.0-75.0) | 158.0 (148.0-165.0) | 70.0 (60.0-84.8) | 27† (NR) |
| Hassan et al. (29) | ||||||
| OA | 77 | 75 | 63.4 (10.3) | 165.0 (8.8) | 83.4 (16.5) | 30.6 (NR) |
| CONT | 63 | 71 | 63.0 (10.7) | 166.0 (7.9) | 68.7 (9.5) | 24.9 (NR) |
| Heiden et al. (30) | ||||||
| OA | 54 | 56 | 65.6 (7.6) | 170.0 (9.0) | 81.4 (14.2) | 28.1 (4.2) |
| CONT | 27 | 67 | 64.2 (5.1) | 170.0 (9.0) | 71.3 (13.8) | 24.4 (3.6) |
| Hurley et al. (31) | ||||||
| OA | 103 | 63 | 60.7 (10.3) | 165.0 (10.2) | 75.9 (14.63) | 28.3 (NR) |
| CONT | 25 | 72 | 65.6 (9.5) | 164.0 (8.5) | 74.0 (16.7) | 27.5 (NR) |
| Kittelson et al. (32) | ||||||
| OA | 17 | 53 | 63.9 (9.0) | 28.3 (5.0) | ||
| CONT | 20 | 50 | 28.3 (11.2) | 25.0 (11.2) | ||
| Lewek et al. (33) | ||||||
| OA | 12 | 42 | 52.6 (7.2) | 31.3 (5.1) | ||
| CONT | 12 | 50 | 48.9 (4.9) | 28.6 (5.6) | ||
| Machner et al. (34) | ||||||
| OA | 18 | 61 | 67.3 (NR) | |||
| Pap et al. (13) | ||||||
| OA | 47 | 77 | 64.0 (5.2) | |||
| CONT | 47 | 77 | 64.0 (5.2) | |||
| Pap et al. (14) | ||||||
| OA | 68 | 60 | 56.7 (9.5) | 164.0 (8.7) | 74.0 (9.4) | 27.5 (NR) |
| OA | 154 | 59 | 65.6 (6.0) | 166.0 (7.2) | 76.0 (11.9) | 25.1 (NR) |
| CONT | 85 | 64 | 58.1 (8.7) | 173.0 (6.8) | 75.0 (9.7) | |
| Petterson et al. (35) | ||||||
| OA | 123 | 54 | 64.9 (8.5) | 170.0 (10.0) | 91.0 (16.4) | 31.4 (4.8) |
| Petterson et al. (36) | ||||||
| OA | 61 | 56 | ||||
| Stevens et al. (37) | ||||||
| OA | 28 | 63.0 (8.8) | ||||
| Suetta et al. (38) | ||||||
| OA | 17 | 29 | ||||
| Tarrago et al. (39) | ||||||
| OA | 21 | 100 | 64.5 (7.7) | 27.5 (5.1) | ||
| CONT | 10 | 100 | 34.1 (11.6) | |||
| Thomas et al. (40) | ||||||
| OA | 22 | 100 | 58.4 (3.0) | 161.2 (5.6) | 84.9 (16.8) | 32.7 (6.6) |
| CONT | 13 | 100 | 57.3 (2.5) | 166.5 (4.7) | 89.2 (13.8) | 31.5 (5.4) |
| Vahtrik et al. (41) | ||||||
| OA | 12 | 100 | 61.0 (6.8) | 33.0 (4.6) | ||
| Ventura et al. (42) | ||||||
| OA | 19 | 53 | 64.0 (6.0) | 171.0 (10.0) | 83.0 (17.0) | 29.0 (5.0) |
*Data presented as mean (s.d.) or as median (range); †Median value.
CONT, control; NR, not reported; OA, osteoarthritis.
QoL, Disability, OA severity scores and pain duration in the participants. Data are presented as mean (s.d.) or as median (range)
| Study/group | KOOS QoL score | Disability score | OA severity, KL score | Pain duration, months |
|---|---|---|---|---|
| Berth et al. (27) | ||||
| OA | Grade 3–4 | |||
| CONT | ||||
| Gapeyeva et al. (28) | ||||
| OA | KKS : 56.0 (20.0–68.0) | Grade 3–4 | 121.2 (NR) | |
| CONT | KKS : 91.0 (65.0–100.0) | |||
| Hassan et al. (29) | ||||
| OA | WOMAC† : 36 (7.0–59.0) | |||
| CONT | ||||
| Heiden et al. (30) | ||||
| OA | 33.7 (15.8) | KOOS-ADLs: 60.0 (20.0) | ||
| CONT | 88.4 (15.3) | KOOS-ADLs: 96.6 (5.7) | ||
| Hurley et al. (31) | ||||
| OA | MLI: 11 (10.42–12.08)* | |||
| CONT | MLI: 1 (0-3)* | |||
| Kittelson et al. (32) | ||||
| OA | WOMAC: NR | |||
| CONT | WOMAC: NR | |||
| Lewek et al. (33) | ||||
| OA | ||||
| CONT | ||||
| Machner et al. (34) | ||||
| OA | Grade 3 | |||
| Pap et al. (13) | ||||
| OA | Grade 4 | |||
| CONT | ||||
| Pap et al. (14) | ||||
| OA | WOMAC: 5.2 (1.6) | Grade 2 | 34 (NR) | |
| OA | WOMAC: 5.5 (1.7) | Grade 4 | 54 (NR) | |
| CONT | WOMAC: 0 (0) | |||
| Petterson et al. (35) | ||||
| OA | ||||
| Petterson et al. (36) | ||||
| OA | Grade 4 | |||
| Stevens et al. (37) | ||||
| OA | KOOS-ADLs: 56.0 (18.0) | |||
| Suetta et al. (38) | ||||
| OA | Grades 2–4 | |||
| Tarrago et al. (39) | ||||
| OA | WOMAC: 57.9 (13.2) | Grades 3–4 | ||
| CONT | ||||
| Thomas et al. (40) | ||||
| OA | WOMAC: 42.5 (18.1) | Grades 2–4 | ||
| CONT | WOMAC: 36.46 (15.26) | Grades 0–1 | ||
| Vahtrik et al. (41) | ||||
| OA | Grades 3–4 | |||
| Ventura et al. (42) | ||||
| OA | 29.0 (17.0) | KOOS-ADLs: 66.0 (18.0) |
*Values in parentheses are 95% CI; †Disability score.
ADL, activity daily living; CONT, control; KL, Kellgren–Lawrence; KSS, Knee Society Score; KOOS, Knee injury and Osteoarthritis Outcome Score; MLI, median Lequesne index; OA, osteoarthritis; QoL, quality of life; WOMAC, Western Ontario and McMaster Universities Arthritis Index.
Transcranial magnetic stimulation
Two studies assessed differences in TMS parameters (specifically ICF, RMT, and SICI) between knee OA participants (n = 38) and healthy controls (n = 30). No studies assessed AMT or CMEPs.
Intracortical facilitation
No differences were observed between knee OA cases and healthy controls (Fig. 2A); however, a non-significant moderate effect towards reduced ICF in OA cases was observed (s.m.d. = −0.43, 95% CI = −0.92 to 0.07, P = 0.09).
(A) Intracortical facilitation comparison between OA cases and healthy controls. (B) Resting motor threshold comparison between OA cases and healthy controls. (C) Short-interval intracortical inhibition comparison between OA cases and healthy controls.
Citation: EFORT Open Reviews 8, 12; 10.1530/EOR-23-0092
Resting motor threshold
No differences were observed between knee OA cases and healthy controls (Fig. 2B); however, a non-significant small effect towards an increased RMT in OA cases was observed (s.m.d. = 0.33, 95% CI = −0.33–0.99, P = 0.33), driven by a large effect from Tarrago et al. (2016) (39) which assessed upper limb.
Short-interval intracortical inhibition
No differences in SICI were observed between knee OA cases and healthy controls (Fig. 2C), with a non-significant, negligible effect.
Motor evoked potentials
One study assessed differences in MEP, with no differences between knee OA and healthy controls observed (39).
Silent period
One study assessed differences in SP, with a reduced silent period in OA compared to controls (39).
Voluntary activation of the quadriceps
Comparison between OA group and healthy controls
All studies comparing voluntary activation between OA participants (n = 567) and controls (n = 386) were performed in knee OA. Significant differences were seen between groups (Fig. 3), with OA cases having reduced voluntary activation compared to controls (s.m.d. = −0.84, 95% CI = −1.12 to −0.56, P < 0.001).
Voluntary activation comparison between OA limb and healthy controls.
Citation: EFORT Open Reviews 8, 12; 10.1530/EOR-23-0092
Comparison between limbs
Eight studies (n = 321) compared voluntary activation of the quadriceps between knee OA cases symptomatic/most symptomatic limb and non/less symptomatic limb, and one study examined hip OA (n = 17). Significant differences were seen between limbs (Fig. 4), with the symptomatic/most symptomatic limb having reduced voluntary activation (s.m.d. = −0.42, 95% CI = −0.75 to −0.09, P = 0.01). Hip OA appeared to create a larger impairment in voluntary activation (s.m.d. = −0.71) compared to knee OA (s.m.d. = −0.30); however, the confidence intervals are broad, as only one study examines hip OA.
Comparison of voluntary activation of the quadriceps between painful/most-painful and non/less-painful limb.
Citation: EFORT Open Reviews 8, 12; 10.1530/EOR-23-0092
Assessment of clinical diversity and statistical heterogeneity
Whilst 15/18 (83%) studies did not report all demographic data, studies included cohorts which were representative of the general OA community. Most studies included both sexes and varying ages, BMI, and disability levels (Table 2). Therefore, generalisability is not a limitation of the study.
Heterogeneity (I 2 = 70%) was observed in the meta-analysis assessing voluntary activation between OA limb and healthy controls. For all other meta-analyses, when sub-grouping by the affected joints (hip or knee), statistical heterogeneity was not observed.
Assessment of quality and bias in included studies
The overall quality for each study was assessed as low (Appendix H). All studies scored low quality on at least two of the eight quality domains. Eleven studies (61%) had a sample of ≤50 OA cases (27, 28, 32, 33, 34, 37, 38, 39, 40, 41, 42), with the remaining studies having samples of <200 OA cases (13, 14, 29, 30, 31, 35, 36). Publication bias did not appear to be present on visual inspection of funnel plots of studies from the analyses of voluntary activation.
Assessment of the certainty of the body of evidence
Transcranial magnetic stimulation between OA cases and healthy controls
The certainty of the body of the evidence was reported as ‘very low’. Only two studies were included, with limited TMS-based measures assessed. The number of OA cases in both studies was fewer than 25 participants, presenting wide confidence intervals, so the evidence was downgraded twice forimprecision. The methodological quality for both studies was recorded as low; therefore, the certainty of the evidence was downgraded for risk of bias. Due to a substantial I 2 statistic the certainty of the evidence was downgraded for inconsistency. Indirectness and publication bias were not judged as reasons to downgrade the evidence.
Quadriceps voluntary activation comparison between OA cases and healthy controls
The certainty of the body of the evidence was reported as ‘low’. The methodological quality for all studies was recorded as low; therefore, the certainty of the evidence was downgraded for risk of bias. Due to a substantial I 2 statistic the certainty of the evidence was downgraded for inconsistency. Indirectness, imprecision, and publication bias were not judged as reasons to downgrade the evidence.
Quadriceps voluntary activation comparison between OA cases symptomatic/most symptomatic limb and non/less symptomatic limb
The certainty of the body of the evidence was reported as ‘moderate’. The methodological quality for all studies was recorded as low; therefore, the certainty of the evidence was downgraded for risk of bias. Inconsistency, indirectness, imprecision, and publication bias were not judged as reasons to downgrade the evidence.
Discussion
Our systematic review and meta-analysis comprehensively synthesised the evidence regarding motor cortex descending drive and voluntary activation for people with lower-limb OA. The majority of included studies assessed voluntary activation (94%) with only two studies using TMS (11%). Our findings indicate that: people with knee OA had large impairments in voluntary activation of their quadriceps muscle compared to healthy controls; people with hip and knee OA have moderate impairments in voluntary activation of the quadriceps on their more painful limb compared to their other limb; trends suggest that cortical excitability/inhibition assessed using TMS may differ in those with knee OA compared to controls, but these findings were not statistically significant. A single study found reduced SP in knee OA compared to controls. Limitations of the current body of evidence and methodological considerations regarding neural assessment are also discussed.
The contributions of motor cortex descending drive to voluntary activation deficits in people with lower-limb OA are still largely unknown
Whilst voluntary activation of the quadriceps was found to be impaired in people with hip and knee OA, cortical excitability and inhibition (one of the mechanisms that might contribute to impaired voluntary activation) did not significantly differ between people with and without OA in our review. However, since this judgement is based on a low number of included studies at a high risk of bias, and hence very low-certainty evidence, our results are likely to be substantially different than the true effect size. Beyond the negative influence of imprecision, inconsistency, and risk of bias on the certainty of the evidence, two significant methodological concerns need highlighting when interpreting the results of the included studies that assessed TMS. First, Tarrago et al. (2016) (39) recorded responses from the hand muscle, not the limb directly impacted by OA (e.g. quadriceps), meaning all data collected were from TMS assessment to the regions of the cortex responsible for upper limb motor responses. Second, Kittelson et al. (2014) (32) placed electrodes over the vastus lateralis muscle only, rather than all components of the quadriceps muscle group, with muscles other than vastus lateralis also playing an important functional role in hip and knee OA (43). Therefore, whilst some preliminary studies have not shown a significant difference in RMT, ICF, and SICI between knee OA and healthy controls, there is little certainty in these outcomes and no studies have explored other variables such as AMT or CMEP. Finally, both included studies were performed in knee OA, with no studies examining hip or ankle OA, meaning motor cortex responses implicated in descending motor drive in people with these conditions are unknown.
Voluntary activation of the quadriceps is impaired in people with lower-limb OA
Our review found that voluntary activation of the quadriceps was impaired (large effect size) in people with knee OA compared to healthy controls (SMD = −0.84, 95% CI = −1.12 to −0.56). Furthermore, voluntary activation of the more symptomatic limb was impaired (moderate effect size) compared to the other limb in people with hip and knee OA (SMD = −0.32, 95% CI = −0.47 to −0.16). These findings provide guidance for clinicians to specifically target voluntary quadriceps activation in exercise rehabilitation programs. Only one study compared quadriceps voluntary activation between limbs in people with hip OA, with all other studies examining knee OA. This represents a significant gap in the literature, especially considering that hip OA is the fastest growing body region for joint OA in terms of prevalence (44). Furthermore, the considerable societal and healthcare burden of hip OA is projected to rise exponentially in the near future (45). It cannot be assumed that knee OA findings are generalisable to other regions of lower-limb OA and based on the findings of our meta-analysis, the extent of quadriceps inhibition on the symptomatic side appeared greater for participants with hip OA than knee OA (Fig. 4). Therefore, an urgent need is required for future research to investigate whether similar impairments in voluntary activation exist in other regions of OA, such as the hip, ankle, and foot.
Methodological recommendations for future research
Methodological quality of future studies assessing TMS and voluntary activation in lower-limb OA could be improved with the use of more robust outcome assessments. Only five studies (28%) reported potentially confounding variables (30, 32, 33, 38, 39), and only two studies (11%) actively implemented strategies to then deal with variables that may influence TMS or voluntary activation (30, 33). Therefore, we provide some suggestions for future research. First, although basing TMS parameters off motor threshold is accepted and appropriate for some measures, it may not be the most suitable approach for paired-pulse outcomes where a set conditioning and test pulse intensity can produce large heterogeneity of responses amongst individuals. Using a test pulse that elicits a common MEP amplitude (e.g. 1 mV or even 0.5 mV for quadriceps) may offer greater normalisation as levels of inhibition and facilitation can be affected by the size of the test MEP. Additionally, although common to set a standard conditioning pulse intensity (e.g. 70–80% of motor threshold), this approach may not provide the most sensitive information. An alternative approach would be to determine the maximal amount of inhibition or facilitation (e.g. SICI and ICF) achievable in each individual by assessing a range of conditioning stimulus intensities (46). As for voluntary activation, it is evident that many studies did not use (or at least report) muscle/nerve stimulation intensities that were supramaximal (how data were normalised was also infrequently reported). In addition to this, the same intensity was often applied for all individuals. This presents a problem for the true assessment of voluntary activation levels as submaximal stimulation intensities may underestimate or overestimate voluntary activation as not all muscle fibres are excited during stimulation. A common stimulus intensity across participants (or between limbs) is also problematic as electrically evoked activation of the muscle tissue under the electrodes is likely to be affected by muscle and fat mass, skin resistance, and deconditioning. We also note that no research has examined cervicomedullary stimulation, to delineate spinal from supraspinal activity. This may provide even more focal ability for determining the origin of any central nervous system changes in OA. Our systematic review also demonstrates poor reporting of baseline characteristics in most studies as well as clear differences between OA and control groups in some baseline characteristics (e.g. age). This is a problem as age (amongst other variables such as certain medications) can significantly impact voluntary activation (47) and TMS (48) outcomes and should be controlled for in future studies.
Conclusion
Low-certainty evidence demonstrated people with knee OA had substantial impairments in voluntary activation of their quadriceps muscle when compared to healthy controls, with no studies comparing people with other lower-limb OA regions (or other muscle groups). With moderate certainty we conclude that people with hip and knee OA have larger impairments in voluntary activation of the quadriceps in their more painful limb compared to their non/other limb. Finally, whilst trends are demonstrated, very low-certainty evidence was found that cortical excitability/inhibition assessed using TMS did not significantly differ between knee OA cases and controls, meaning the cause of voluntary activation deficits are still unknown. Methodological quality (assessment techniques and data synthesis) of the studies requires improvement to increase certainty of findings and assist clinicians managing people with OA.
Supplementary materials
This is linked to the online version of the paper at https://doi.org/10.1530/EOR-23-0092.
ICMJE Conflict of Interest Statement
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding Statement
Support for this study was provided by an Edith Cowan University Early-Mid Career Researcher, Stream 2 (Pilot Project) Grant 2022 (G1006474_SUB/108223) Grant. MCM is supported by a Department of Health WA Near-Miss Award (G1006605). Study funders had no involvement within the systematic review and meta-analysis.
Author contribution statement
MCM conceived the study and MCM, CL, EKR, and ABM designed the study. MCM conducted the search, MCM and WH screened studies, SM translated and performed data extraction for non-English studies. MCM and CL performed quality appraisal, and MCM and CS performed data extraction. All authors contributed to interpretation of data analysis and writing the final manuscript.
Acknowledgements
The authors would like to acknowledge Mr Matthieu Janin for assistance in translation of the articles.
References
- 1.↑
GBD 2016 Disease and Injury Incidence and Prevalence Collaborators. Global, regional, and national incidence, prevalence, and years lived with disability for 328 diseases and injuries for 195 countries, 1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. Lancet 2017 390 1211–1259. (https://doi.org/10.1016/S0140-6736(1732154-2)
- 2.↑
Australian Institute of Health and Welfare. Osteoarthritis https://www.aihw.gov.au/reports/chronic-musculoskeletal-conditions/osteoarthritis: Cat. no. PHE 232. Canberra: AIHW 2020.
- 3.↑
Cook MJ, Lunt M, Board T, & O'Neill TW. The impact of frailty on short-term mortality following primary total hip and knee arthroplasty due to osteoarthritis. Age and Ageing 2022 51. (https://doi.org/10.1093/ageing/afac118)
- 4.↑
Laires PA, Canhão H, Rodrigues AM, Eusébio M, Gouveia M, & Branco JC. The impact of osteoarthritis on early exit from work: results from a population-based study. BMC Public Health 2018 18 472. (https://doi.org/10.1186/s12889-018-5381-1)
- 5.↑
de Melo LRS, Hunter D, Fortington L, Peeters A, Seward H, Vertullo C, Hills AP, Brown W, Feng Y, & Lloyd DG. National osteoarthritis Strategy brief report: prevention of osteoarthritis. Australian Journal of General Practice 2020 49 272–275. (https://doi.org/10.31128/AJGP-08-19-5051-01)
- 6.↑
Arant KR, Katz JN, & Neogi T. Quantitative sensory testing: identifying pain characteristics in patients with osteoarthritis. Osteoarthritis and Cartilage 2022 30 17–31. (https://doi.org/10.1016/j.joca.2021.09.011)
- 7.↑
Heerey JJ, Kemp JL, Mosler AB, Jones DM, Pizzari T, Souza RB, & Crossley KM. What is the prevalence of imaging-defined intra-articular hip pathologies in people with and without pain? A systematic review and meta-analysis. British Journal of Sports Medicine 2018 52 581–593. (https://doi.org/10.1136/bjsports-2017-098264)
- 8.↑
Ghouri A, Muzumdar S, Barr AJ, Robinson E, Murdoch C, Kingsbury SR, & Conaghan PG. The relationship between meniscal pathologies, cartilage loss, joint replacement and pain in knee osteoarthritis: a systematic review. Osteoarthritis and Cartilage 2022 30 1287–1327. (https://doi.org/10.1016/j.joca.2022.08.002)
- 9.↑
Murphy M, Travers MJ, Gibson W, Chivers P, Debenham J, Docking S, & Rio E. The rate of improvement of pain and function in mid-portion Achilles Tendinopathy with loading protocols: a systematic review and longitudinal meta-analysis. Sports Medicine . 2018 48 1875–1891. (https://doi.org/10.1007/s40279-018-0932-2)
- 10.↑
Skou ST, & Roos EM. Good Life with osteoarthritis in Denmark (GLA:D™): evidence-based education and supervised neuromuscular exercise delivered by certified physiotherapists nationwide. BMC Musculoskeletal Disorders 2017 18 72. (https://doi.org/10.1186/s12891-017-1439-y)
- 11.↑
Vincent KR, & Vincent HK. Resistance exercise for knee osteoarthritis. PM and R 2012 4(5) S45–S52. (https://doi.org/10.1016/j.pmrj.2012.01.019)
- 12.↑
Murphy MC, Travers M, Chivers P, Debenham J, Docking SI, Rio EK, Gibson W, & Ardern C. Can we really say getting stronger makes your tendon feel better? No current evidence of a relationship between change in Achilles tendinopathy pain or disability and changes in triceps surae structure or function when completing rehabilitation: a systematic review. Journal of Science and Medicine in Sport 2023 26 253–260. (https://doi.org/10.1016/j.jsams.2023.03.007)
- 13.↑
Pap G, Machner A, & Awiszus F. Functional changes of the quadriceps femoris muscle in patients with osteoarthritis of the knee. Zeitschrift fur Rheumatologie 2000 59 380–387. (https://doi.org/10.1007/s003930070046)
- 14.↑
Pap G, Machner A, & Awiszus F. Strength and voluntary activation of the quadriceps femoris muscle at different severities of osteoarthritic knee joint damage. Journal of Orthopaedic Research 2004 22 96–103. (https://doi.org/10.1016/S0736-0266(0300128-1)
- 15.↑
Chang WJ, Bennell KL, Hodges PW, Hinman RS, Young CL, Buscemi V, Liston MB, & Schabrun SM. Addition of transcranial direct current stimulation to quadriceps strengthening exercise in knee osteoarthritis: a pilot randomised controlled trial. PLoS One 2017 12 e0180328. (https://doi.org/10.1371/journal.pone.0180328)
- 16.↑
Simis M, Imamura M, de Melo PS, Marduy A, Pacheco-Barrios K, Teixeira PEP, Battistella L, & Fregni F. Increased motor cortex inhibition as a marker of compensation to chronic pain in knee osteoarthritis. Scientific Reports 2021 11 24011. (https://doi.org/10.1038/s41598-021-03281-0)
- 17.↑
Parker RS, Lewis GN, Rice DA, & McNair PJ. Is motor cortical excitability altered in people with chronic pain? A systematic review and meta-analysis. Brain Stimulation 2016 9 488–500. (https://doi.org/10.1016/j.brs.2016.03.020)
- 18.↑
Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, Shamseer L, Tetzlaff JM, Akl EA, Brennan SE, et al.The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ 2021 372 n71. (https://doi.org/10.1136/bmj.n71)
- 19.↑
Travers MJ, Murphy MC, Debenham JR, Chivers P, Bulsara MK, Bagg MK, Palsson TS, & Gibson W. Should this systematic review and meta-analysis change my practice? Part 1: exploring treatment effect and trustworthiness. British Journal of Sports Medicine 2019 53 1488–1492. (https://doi.org/10.1136/bjsports-2018-099958)
- 20.↑
Travers MJ, Murphy MC, Debenham JR, Chivers P, Bulsara MK, Bagg MK, Palsson TS, & Gibson W. Should this systematic review and meta-analysis change my practice? Part 2: exploring the role of the comparator, diversity, risk of bias and confidence. British Journal of Sports Medicine 2019 53 1493–1497. (https://doi.org/10.1136/bjsports-2018-099959)
- 21.↑
Petersson IF, Boegård T, Saxne T, Silman AJ, & Svensson B. Radiographic osteoarthritis of the knee classified by the Ahlbäck and Kellgren & Lawrence systems for the tibiofemoral joint in people aged 35–54 years with chronic knee pain. Annals of the Rheumatic Diseases 1997 56 493–496. (https://doi.org/10.1136/ard.56.8.493)
- 22.↑
The Cochrane Collaboration. Cochrane handbook for systematic reviews of interventions. Available at: https://training.cochrane.org/handbook/current2022
- 23.↑
Dechartres A, Trinquart L, Boutron I, & Ravaud P. Influence of trial sample size on treatment effect estimates: meta-epidemiological study. BMJ 2013 346 f2304. (https://doi.org/10.1136/bmj.f2304)
- 24.↑
Murphy MC, George H, Naqi M, Owen PJ, Chivers P, & Hart NH. Musculoskeletal injury epidemiology in law enforcement and firefighter recruits during physical training: a systematic review. BMJ Open Sport and Exercise Medicine 2022 8. (https://doi.org/10.1136/bmjsem-2021-001289)
- 25.↑
Murphy MC, Travers MJ, Chivers P, Debenham JR, Docking SI, Rio EK, & Gibson W. Efficacy of heavy eccentric calf training for treating mid-portion Achilles tendinopathy: a systematic review and meta-analysis. British Journal of Sports Medicine 2019 53 1070–1077. (https://doi.org/10.1136/bjsports-2018-099934)
- 26.↑
Higgins JPT, & Green S. Cochrane Handbook for Systematic Reviews of Interventions, Version 5.1.0 [Updated March 2011]. Available at: www.handbook.cochrane.org2011
- 27.↑
Berth A, Urbach D, & Awiszus F. Improvement of voluntary quadriceps muscle activation after total knee arthroplasty. Archives of Physical Medicine and Rehabilitation 2002 83 1432–1436. (https://doi.org/10.1053/apmr.2002.34829)
- 28.↑
Gapeyeva H, Buht N, Peterson K, Ereline J, Haviko T, & Pääsuke M. Quadriceps femoris muscle voluntary isometric force production and relaxation characteristics before and 6 months after unilateral total knee arthroplasty in women. Knee Surgery, Sports Traumatology, Arthroscopy 2007 15 202–211. (https://doi.org/10.1007/s00167-006-0166-y)
- 29.↑
Hassan BS, Mockett S, & Doherty M. Static postural sway, proprioception, and maximal voluntary quadriceps contraction in patients with knee osteoarthritis and normal control subjects. Annals of the Rheumatic Diseases 2001 60 612–618. (https://doi.org/10.1136/ard.60.6.612)
- 30.↑
Heiden TL, Lloyd DG, & Ackland TR. Knee extension and flexion weakness in people with knee osteoarthritis: is antagonist cocontraction a factor? Journal of Orthopaedic and Sports Physical Therapy 2009 39 807–815. (https://doi.org/10.2519/jospt.2009.3079)
- 31.↑
Hurley MV, Scott DL, Rees J, & Newham DJ. Sensorimotor changes and functional performance in patients with knee osteoarthritis. Annals of the Rheumatic Diseases 1997 56 641–648. (https://doi.org/10.1136/ard.56.11.641)
- 32.↑
Kittelson AJ, Thomas AC, Kluger BM, & Stevens-Lapsley JE. Corticospinal and intracortical excitability of the quadriceps in patients with knee osteoarthritis. Experimental Brain Research 2014 232 3991–3999. (https://doi.org/10.1007/s00221-014-4079-6)
- 33.↑
Lewek MD, Rudolph KS, & Snyder-Mackler L. Quadriceps femoris muscle weakness and activation failure in patients with symptomatic knee osteoarthritis. Journal of Orthopaedic Research 2004 22 110–115. (https://doi.org/10.1016/S0736-0266(0300154-2)
- 34.↑
Machner A, Pap G, & Awiszus F. Evaluation of quadriceps strength and voluntary activation after unicompartmental arthroplasty for medial osteoarthritis of the knee. Journal of Orthopaedic Research 2002 20 108–111. (https://doi.org/10.1016/S0736-0266(0100068-7)
- 35.↑
Petterson SC, Barrance P, Buchanan T, Binder-Macleod S, & Snyder-Mackler L. Mechanisms underlying quadriceps weakness in knee osteoarthritis. Medicine and Science in Sports and Exercise 2008 40 422–427. (https://doi.org/10.1249/MSS.0b013e31815ef285)
- 36.↑
Petterson SC, Barrance P, Marmon AR, Handling T, Buchanan TS, & Snyder-Mackler L. LTime course of quad strength, area, and activation after knee arthroplasty and strength training. Medicine and Science in Sports and Exercise 2011 43 225–231. (https://doi.org/10.1249/MSS.0b013e3181eb639a)
- 37.↑
Stevens JE, Mizner RL, & Snyder-Mackler L. Quadriceps strength and volitional activation before and after total knee arthroplasty for osteoarthritis. Journal of Orthopaedic Research 2003 21 775–779. (https://doi.org/10.1016/S0736-0266(0300052-4)
- 38.↑
Suetta C, Aagaard P, Magnusson SP, Andersen LL, Sipilä S, Rosted A, Jakobsen AK, Duus B, & Kjaer M. Muscle size, neuromuscular activation, and rapid force characteristics in elderly men and women: effects of unilateral long-term disuse due to hip-osteoarthritis. Journal of Applied Physiology 2007 102 942–948. (https://doi.org/10.1152/japplphysiol.00067.2006)
- 39.↑
Tarragó MDGL, Deitos A, Brietzke AP, Vercelino R, Torres ILS, Fregni F, & Caumo W. Descending control of nociceptive processing in knee osteoarthritis is associated with intracortical disinhibition: an exploratory study. Medicine (Baltimore) 2016 95 e3353. (https://doi.org/10.1097/MD.0000000000003353)
- 40.↑
Thomas AC, Sowers M, Karvonen-Gutierrez C, & Palmieri-Smith RM. Lack of quadriceps dysfunction in women with early knee osteoarthritis. Journal of Orthopaedic Research 2010 28 595–599. (https://doi.org/10.1002/jor.21038)
- 41.↑
Vahtrik D, Gapeyeva H, Aibast H, Ereline J, Kums T, Haviko T, Märtson A, Schneider G, & Pääsuke M. Quadriceps femoris muscle function prior and after total knee arthroplasty in women with knee osteoarthritis. Knee Surgery, Sports Traumatology, Arthroscopy 2012 20 2017–2025. (https://doi.org/10.1007/s00167-011-1808-2)
- 42.↑
Ventura A, Muendle B, Friesenbichler B, Casartelli NC, Kramers I, & Maffiuletti NA. Deficits in rate of torque development are accompanied by activation failure in patients with knee osteoarthritis. Journal of Electromyography and Kinesiology 2019 44 94–100. (https://doi.org/10.1016/j.jelekin.2018.11.010)
- 43.↑
Loureiro A, Constantinou M, Diamond LE, Beck B, & Barrett R. Individuals with mild-to-moderate hip osteoarthritis have lower limb muscle strength and volume deficits. BMC Musculoskeletal Disorders 2018 19 303. (https://doi.org/10.1186/s12891-018-2230-4)
- 44.↑
Ackerman IN, Buchbinder R, & March L. Global Burden of Disease Study 2019: an opportunity to understand the growing prevalence and impact of hip, knee, hand and other osteoarthritis in Australia. International Medical Journal 2022. (https://doi.org/10.1111/imj.15933)
- 45.↑
Ackerman IN, Bohensky MA, Zomer E, Tacey M, Gorelik A, Brand CA, & de Steiger R. The projected burden of primary total knee and hip replacement for osteoarthritis in Australia to the year 2030. BMC Musculoskeletal Disorders 2019 20 90. (https://doi.org/10.1186/s12891-019-2411-9)
- 46.↑
Ruas CV, Taylor JL, Nosaka K, Haff GG, & Latella C. A time-efficient method to determine parameters for measurement of short-interval intracortical inhibition for quadriceps. European Journal of Neuroscience 2020 52 4751–4761. (https://doi.org/10.1111/ejn.14904)
- 47.↑
Rozand V, Sundberg CW, Hunter SK, & Smith AE. Age-related deficits in voluntary activation: a systematic review and meta-analysis. Medicine and Science in Sports and Exercise 2020 52 549–560. (https://doi.org/10.1249/MSS.0000000000002179)
- 48.↑
Bhandari A, Radhu N, Farzan F, Mulsant BH, Rajji TK, Daskalakis ZJ, & Blumberger DM. A meta-analysis of the effects of aging on motor cortex neurophysiology assessed by transcranial magnetic stimulation. Clinical Neurophysiology 2016 127 2834–2845. (https://doi.org/10.1016/j.clinph.2016.05.363)
