Skip to main content
NCBI home page
Wiley Open Access Collection logoLink to Wiley Open Access Collection
. 2025 Aug 25;29(9):e70100. doi: 10.1002/ejp.70100

Effectiveness of Movement Representation Strategies in Knee Pain: A Systematic Review and Meta‐Analyses on Knee Pain and Function

Julie Flebus 1, Javier Martín Núñez 2, María Granados Santiago 3, Alejandro Heredia Ciuró 2, Andrés Calvache Mateo 2, Alba Navas Otero 2, Marie Carmen Valenza 2,✉
PMCID: PMC12376090  PMID: 40852787

ABSTRACT

Introduction

A highly prevalent condition, knee pain often results in significant functional limitations and a reduced quality of life. Due to its multifactorial causes and diverse pathologies, numerous therapeutic approaches have been proposed, each with varying degrees of success. Among these, movement representation strategies have emerged as promising interventions. These techniques engage the central nervous system by using mental simulation of motor actions—such as motor imagery and action observation—without the need for actual physical movement, often focusing on imagining or observing pain‐free, unrestricted motion.

Methods

Because the effectiveness of movement representation strategies in knee pain remains uncertain and no prior synthesis of randomised evidence exists, a systematic review of the literature was conducted for randomised controlled trials indexed from three databases inception to March 2025. Two reviewers performed independent data extraction and methodologic quality assessment of the studies.

Results

Eleven studies were included in this review. The results of pain and function outcomes showed significant improvements after interventions based on movement representation techniques. The meta‐analyses showed that these techniques have a significant effect on pain and function.

Discussion and Conclusion

The results of our review demonstrated notable results from the implementation of movement representation techniques to standard physical therapy aimed at decreasing pain and increasing function in patients with knee pain. The meta‐analyses revealed a significant positive effect of these interventions.

Significance Statement

Movement representation strategies combined with physical therapy have a significant effect on function and pain in patients with knee pain. The most commonly used strategies were motor imagery training and action observation training. The meta‐analyses revealed a significant positive effect of these interventions, showing improvements in both pain and function in patient with knee pain.

Keywords: function, knee, movement representation techniques, pain

1. Introduction

The knee is one of the most commonly affected joints, with up to 15%–50% of older adults experiencing knee pain, which is often accompanied by significant functional impairment (Peat et al. 2001; Bedson et al. 2007; Neogi et al. 2010). In developed countries such as the United States and Switzerland, knee pain ranks among the most prevalent sources of musculoskeletal pain in the adult population, severely impacting mobility and quality of life (Bunt et al. 2018; Turkiewicz et al. 2015). Given the essential role of the lower limbs in weight‐bearing and locomotion, pain and dysfunction in these joints present unique clinical challenges. Therapeutic approaches to knee pain range from conservative management—including physical therapy, pharmacological treatments and lifestyle modifications—to surgical interventions such as arthroscopy or joint replacement (Bennell et al. 2012; Skou et al. 2015).

Despite differences in aetiology, whether related to acute trauma, ligamentous or meniscal injury, cartilage degeneration, or chronic osteoarthritis, there is evidence suggesting a common risk for pain chronification that can persist regardless of conservative or surgical treatment approaches (Vlaeyen et al. 2016; Wylde et al. 2018). Movement representation strategies (MRS) may play a role in counteracting this risk by promoting adaptive cortical reorganisation, modulating sensorimotor processing and reducing the central amplification of pain signals through non‐invasive activation of motor pathways (Hardwick et al. 2018; Paravlic 2022).

Conservative treatments for knee pain encompass a wide range of approaches, including physical therapy, pharmacological management, lifestyle modifications, exercise programmes and modalities, such as electrotherapy and manual therapy. These interventions can be applied in isolation or as complementary strategies alongside surgical treatments to optimise recovery and functional outcomes (Fransen et al. 2015; McAlindon et al. 2014). Among these, MRS have gained increasing attention due to their unique neurophysiological mechanisms. These techniques engage the central nervous system by activating motor‐related brain areas through mental simulation of movement, such as motor imagery, mirror therapy and action observation, without requiring actual physical execution (Grush 2004; Schuster et al. 2011). This mental rehearsal modulates corticospinal excitability, promotes cortical plasticity, and can reduce maladaptive neuroplastic changes associated with pain chronification and motor dysfunction (Page et al. 2015; Lotze and Halsband 2006; Fadiga et al. 1999). Furthermore, these interventions can facilitate motor learning and functional recovery by reinforcing sensorimotor representations and improving motor planning and execution, which is particularly beneficial in conditions where pain limits active movement (Grush 2004; Schuster et al. 2011).

The MRS integrates the CNS in the treatment, since they are therapies based on mental simulation of a motor action without physically performing it, especially through the observation and/or imagination of free movement without pain. In addition to the possibility of combining with the execution of the movement and/or sensory stimulation to facilitate the movement of the joint without pain.

These types of strategies have been used previously in the rehabilitation of low back pain, where positive results have been shown in favour of MRS (Thieme et al. 2016). In relation to knee pain, previous studies have observed benefits of some MRS, such as motor imagery, as well as positive results of MRS on physical and cognitive function in patients with knee and/or hip arthroplasty (Galonski et al. 2023; Riquelme‐Hernández et al. 2022) However, to our knowledge, no one has conducted a systematic review and meta‐analysis of the effectiveness of MRS in the treatment of knee pain and function. Therefore, the aim of our study was to evaluate the effectiveness of MRS on pain and function in patients with knee pain.

2. Methods

2.1. Study Registration

The systematic review was conducted in accordance with PROSPERO (Prospective International Registry of Systematic Reviews) guidelines, obtaining registration number CRD42023478157 to guarantee compliance with public access standards. The review was performed following the principles described in the Preferred Reporting Items for Systematic Reviews and Meta‐Analyses (PRISMA) (Higgins 2011) and in accordance with the intervention guidelines specified in the Cochrane Collaboration manual (Centre for Reviews and Dissemination 2009).

2.2. Search Strategy

A systematic search of the literature was conducted for randomised controlled trials indexed on PubMed, Web of Sciences and Scopus databases from their inception to March 2025 in English. To develop our search strategy, we took into account the inclusion of MeSH terms and keywords used in similar reviews on the subject. The following search strategy was designed for the PubMed database: (‘Mirror Therapy’ OR ‘Mirror Movement Therapy’ OR ‘Mirror Visual Feedback’ OR ‘Mirror Box Therapy’ OR ‘Graded Motor Imagery’ OR ‘movement representation strategies’ OR ‘mental practice’ OR ‘motor imagery’ OR ‘mental imagery’ OR ‘visual imagery’ OR ‘guided imagery’ OR ‘kinesthetic imagery’ OR ‘mental training’ OR ‘movement representation techniques’ OR ‘imagination’ OR ‘action observation’ OR ‘observation training’ OR ‘action observation therapy’) AND (‘knee’ OR ‘knee joint’ OR ‘anterior cruciate ligament’ OR ‘posterior cruciate ligament’ OR ‘knee pain’ OR ‘acute knee pain’ OR ‘chronic knee pain’ OR ‘musculoskeletal knee pain’). Then we adapted this search strategy to the Web of Sciences and Scopus databases. In order to be as accurate as possible, reference lists of relevant reviews related to the terms were reviewed and non‐English language studies were considered for inclusion if translation was possible.

2.3. Study Selection

Studies were systematically selected according to our PICOS strategy (participants, interventions, comparisons, outcome and study design) eligibility criteria: (Andrews et al. 2013) Adults (≥ 18 years) with knee pain regardless of aetiology and/or treatment; (Balshem et al. 2011) Studies were included if they involved the application of movement representation techniques—such as motor imagery, action observation or other cognitively oriented motor simulation approaches—either as stand‐alone interventions or in combination with other conventional treatments (e.g., physical therapy, exercise, or education). This inclusive definition allowed us to capture the shared therapeutic effect of MRTs, irrespective of whether they were applied independently or as part of a multimodal rehabilitation protocol (Bedson et al. 2007). Movement representation strategies had to be compared with usual care or programmes without movement representation strategies; (Bennell et al. 2012) Pain and/or function as an outcome; (Binks et al. 2023) Randomised controlled trial or pilot randomised controlled trials. We did not include abstracts, conference papers and duplicate publications; and articles in languages other than English, French or Spanish.

2.4. Data Extraction

The extracted data included details such as the author, year of publication, age, sample of knee pain aetiology, interventions, pain and function outcomes and characteristics of programmes. Whenever there was insufficient information or ambiguity, we endeavoured to contact the corresponding author of the study via email. If the data remained unclear or communication was unsuccessful, we proceeded with the analysis of the available data. In the same way, data extraction was also carried out independently by two authors.

2.5. Qualitative Analyses

The evaluation reviewed findings and categorised them according to the Grading of Recommendations Assessment, Development and Evaluation (GRADE) system. This framework considers five domains: study design, imprecision, indirectness, inconsistency and publication bias (Guyatt et al. 2008). The available information was divided into four distinct quality categories: (1) Very low quality when analysis of impacts is highly uncertain, with failures in at least three of the five domains; (2) Low quality when further investigation is very likely to significantly impact confidence in the estimated effects and could potentially alter our understanding; (3) Moderate quality when additional research is likely to significantly affect certainty in the impact assessment and may result in changes, typically when one of the five domains fails to meet the criteria; (4) High quality when further research is extremely unlikely to affect confidence in the impact assessment, with all five domains fully meeting the criteria (Andrews et al. 2013).

The assessment of the five domains followed GRADE standards. In the study design domain, studies were downgraded by one level if there was ambiguity or a high risk of bias, along with major limitations in impact assessment. For inconsistency, studies were downgraded by one level if quantitative assessments showed substantial variability between studies, if confidence intervals showed minimal overlap, or if the I 2 statistic indicated large or very large heterogeneity. In the indirectness domain, recommendations were downgraded if there were significant differences in interventions, study populations, or outcomes. For imprecision, studies were downgraded by one level if there were fewer than 400 participants for continuous data (Balshem et al. 2011).

2.6. Risk of Bias of Included Studies

The Cochrane Risk of Bias Assessment Tool (ROB2) (Higgins 2011) was utilised to assess the risk of bias in the included studies. This tool examines seven domains: random sequence generation, allocation concealment, blinding of participants and personnel, blinding of outcome assessment, incomplete outcome data, selective reporting and other potential biases. Each domain within the studies was rated as ‘high risk of bias’, ‘low risk of bias’, or ‘unclear’. Based on the score obtained in each of the items, a final score is established following the same criteria of high, low or some concerns of risk.

2.7. Meta‐Analyses

Review Manager software (RevMan version 5.1, updated March 2011) was employed to conduct a meta‐analysis on all randomised controlled trials (RCTs) that assessed pain and function. The sample sizes, means, standard mean differences and standard deviations (SDs) for each post‐intervention variable were input into the software. In cases where means and standard deviations were not reported, the authors were contacted. Variables expressed in non‐comparable units were ultimately excluded from the meta‐analysis.

When standard deviations were missing but p‐values or 95% confidence intervals were provided, these were calculated using the Review Manager calculator. If studies employed different measurement tools, the standard mean difference (SMD) was chosen as the measure of effect size.

Overall mean effect sizes were estimated employing random effect models or fixed effect models based on the degree of statistical heterogeneity (fixed effect models were used for heterogeneity sizes less than 50%). The effects were expressed as mean differences (MDs) and standard mean differences (SMDs) with corresponding confidence intervals. Additionally, forest plots were visually inspected for outlier studies. We analysed forest plots for outlier studies, explored sources of heterogeneity and performed sensitivity analyses excluding assays with high risk of detection bias or attrition.

3. Results

An initial search of the electronic databases obtained 546 records. After removing duplicates, 515 studies were screened. Studies that did not meet the inclusion criteria defined by the PICOS strategy were excluded. Title and abstract‐based screening resulted in 77 articles being selected. Only five studies were not available as full‐text articles and were subsequently evaluated against the inclusion criteria. After full‐text screening, 11 studies were included. We examined the reference lists of pertinent studies to determine if they contained mentions of additional studies that could meet the criteria for inclusion in the review, resulting in the identification of one study through this approach. Finally, 12 studies were included in our systematic review (Cupal and Brewer 2001; Lebon, Byblow, et al. 2012; Lebon, Guillot, et al. 2012; Park et al. 2014; Wilczynska et al. 2015; Villafañe et al. 2017; Moukarzel, Di Rienzo, et al. 2019; Moukarzel, Guillot, et al. 2019; Paravlic et al. 2019; Briones‐Cantero et al. 2020; Öztürk et al. 2021; Candiri et al. 2023). The PRISMA flow diagram illustrating the study selection process is shown in Figure 1.

FIGURE 1.

FIGURE 1

Open in a new tab

Flow chart of the selected studies.

3.1. Study Characteristics

Of the 11 included studies, 8 studies (Cupal and Brewer 2001; Lebon, Byblow, et al. 2012; Lebon, Guillot, et al. 2012; Park et al. 2014; Wilczynska et al. 2015; Paravlic et al. 2019; Briones‐Cantero et al. 2020; Öztürk et al. 2021; Candiri et al. 2023) were randomised controlled trials and 3 were pilot studies (Villafañe et al. 2017; Moukarzel, Di Rienzo, et al. 2019; Moukarzel, Guillot, et al. 2019) of randomised controlled trials. A total sample of 255 individuals with knee pain, ranging in age from 18.2 ± 8.2 to 72.5 ± 5.5, was included. Regarding the sex of the total sample included, 85 men and 144 women were included in our review; however, the studies by Park et al. and Wilczynska et al. (2015) did not provide this information. Regarding the aetiology of knee pain present in these individuals, to highlight that seven studies included patients with knee arthroplasty (Park et al. 2014; Villafañe et al. 2017; Moukarzel, Di Rienzo, et al. 2019; Moukarzel, Guillot, et al. 2019; Paravlic et al. 2019; Briones‐Cantero et al. 2020; Candiri et al. 2023) 2 ACL reconstructions (Cupal and Brewer 2001; Lebon, Byblow, et al. 2012; Lebon, Guillot, et al. 2012) while only one study included individuals with arthroscopy (Wilczynska et al. 2015) and another with osteoarthritis (Öztürk et al. 2021; Table 1).

TABLE 1.

Characteristics of studies.

Studies Type of study Sample Sample Age (years ± SD) Gender (M/W) Aetiology
Cupal and Brewer (2001) RCT 30 18.2 ± 8.2 16/14 ACL reconstruction
Lebon, Byblow, et al. (2012) RCT 12 28.5 ± 5 10/2 ACL reconstruction
Park et al. (2014) RCT 18 71.6 ± 11.6 NR Arthroplasty
Wilczynska et al. (2015) RCT 10 35 NR Arthroscopy
Villafañe et al. (2017) Pilot RCT 31 70.3 ± 7.6 10/23 Arthroplasty
Moukarzel, Di Rienzo, et al. (2019) Pilot RCT 20 69.6 ± 3.25 4/16 Arthroplasty
Paravlic et al. (2019) RCT 34 61.1 ± 5.3 19/15 Arthroplasty
Moukarzel, Di Rienzo, et al. (2019), Moukarzel, Guillot, et al. (2019) Pilot RCT 24 70 ± 2.89 4/20 Arthroplasty
Briones‐Cantero et al. (2020) RCT 24 72.5 ± 5.5 15/9 Arthroplasty
Öztürk et al. (2021) RCT 34 59.7 ± 6.6 5/29 Osteoarthritis
Candiri et al. (2023) RCT 18 66 ± 5.6 2/16 Arthroplasty
Open in a new tab

3.2. Risk of Bias

We evaluated the risk of bias in the randomised controlled trials and pilot studies included in our systematic review using the Cochrane Risk of Bias Tool. Among the studies reviewed, 4 were found to have a low risk of bias (Villafañe et al. 2017; Moukarzel, Di Rienzo, et al. 2019; Moukarzel, Guillot, et al. 2019; Briones‐Cantero et al. 2020; Candiri et al. 2023) while six others presented some concerns (Cupal and Brewer 2001; Lebon, Byblow, et al. 2012; Lebon, Guillot, et al. 2012; Park et al. 2014; Wilczynska et al. 2015; Paravlic et al. 2019; Öztürk et al. 2021; Figure 2).

FIGURE 2.

FIGURE 2

Open in a new tab

Risk of bias.

3.3. Results of Studies

All the studies included in our review performed movement representation strategies in combination with standard physical therapy; however, different strategies were used. Six studies performed a motor imagery training strategy (Lebon, Byblow, et al. 2012; Lebon, Guillot, et al. 2012; Moukarzel, Di Rienzo, et al. 2019; Moukarzel, Guillot, et al. 2019; Paravlic et al. 2019; Briones‐Cantero et al. 2020; Candiri et al. 2023), three studies performed action observation training (Park et al. 2014; Villafañe et al. 2017; Öztürk et al. 2021) while the study of Wilczynska et al. (2015) performed visualisation and the study of Cupal and Brewer (2001) combined guided imagery with relaxation. As for the control group, all studies performed standard physiotherapy. In relation to the characteristics of the programmes, the weeks of intervention ranged between 2 and 24 weeks, the frequency of intervention per week between 3 and 5, taking into account the peculiarity of the study of Cupal and Brewer (2001) where the participants received the intervention either weekly or every 2 weeks. Finally, for the dose of the session, it ranged between 30 and 60 min, where it is necessary to highlight that the study of Paravlic et al. (2019) only provided information on the time that the motor imagery training was performed (15 min) but not the total of the session.

In relation to the outcomes and their results, eight studies evaluated pain (Cupal and Brewer 2001; Lebon, Byblow, et al. 2012; Lebon, Guillot, et al. 2012; Park et al. 2014; Wilczynska et al. 2015; Villafañe et al. 2017; Moukarzel, Di Rienzo, et al. 2019; Moukarzel, Guillot, et al. 2019; Briones‐Cantero et al. 2020; Öztürk et al. 2021; Candiri et al. 2023) all of which used the VAS tool, except for the study by Wilczynska et al. (2015) which evaluated it using the Latin Pain Scale. As for function, this was evaluated in nine studies (Park et al. 2014; Villafañe et al. 2017; Moukarzel, Di Rienzo, et al. 2019; Moukarzel, Guillot, et al. 2019; Paravlic et al. 2019; Briones‐Cantero et al. 2020; Öztürk et al. 2021; Candiri et al. 2023) where more heterogeneity was found in the tools used (WOMAC; TUG; Barthel Index; Chair sit‐to‐stand). The results of these outcomes showed significant improvements in pain after the intervention both in comparison with baseline measurements and in the control group in 7 of the 9 studies (Cupal and Brewer 2001; Park et al. 2014; Wilczynska et al. 2015; Villafañe et al. 2017; Moukarzel, Di Rienzo, et al. 2019; Moukarzel, Guillot, et al. 2019; Briones‐Cantero et al. 2020; Candiri et al. 2023; Table 2).

TABLE 2.

Characteristics of interventions.

Studies Intervention Programme (week); frequency (days/week); dose (min of session) Outcomes Results
Cupal and Brewer (2001)

EG1: Guided imagery + relaxation + physical therapy

CG (1): Physical therapy + visual stimulus

CG (2): Physical therapy

 24; 0–1; 40–60 Pain (EVA)

Pre > Post (p < 0.05)

EG > CG (1,2) (p < 0.05)
Lebon, Byblow, et al. (2012)

EG: Motor imagery training + physical therapy

CG: Physical therapy

 4; 3; 45 Pain (EVA)

Pre > Post (p < 0.05)

EG < CG (NS)
Park et al. (2014)

EG: Action observation training + physical therapy

CG: Physical therapy

 3; 3; 40 Pain (EVA); Function (WOMAC)

Pain:

Pre > Post (p < 0.05)

EG < CG (p < 0.05)

Function:

Pre < Post (p < 0.001)

EG > CG (p < 0.001)
Wilczynska et al. (2015)

EG: Visualisation + physical therapy

CG: Physical therapy

 4; 3–4; NR Pain (LPS)

Pre > Post (p < 0.05)

EG < CG (p < 0.05)
Villafañe et al. (2017)

EG: Action observation training + physical therapy

CG: Physical therapy

 2; 5; 60 Pain (EVA); Function (Barthel index)

Pain:

Pre > Post (p < 0.05)

EG < CG (p < 0.05)

Function:

Pre < Post (p < 0.05)

EG > CG (p < 0.05)
Moukarzel, Di Rienzo, et al. (2019)

EG: Motor imagery training + physical therapy

CG: Physical therapy

 NR Pain (EVA); Function (TUG)

Pain:

Pre > Post (p < 0.001)

EG < CG (p < 0.001)

Function:

Pre < Post (p < 0.001)

EG > CG (p < 0.001)
Paravlic et al. (2019)

EG: Motor imagery training + physical therapy

CG: Physical therapy

 NR; 5; 15 (MI) Function (Chair sit‐to‐stand)

Pre < Post (p < 0.001)

EG > CG (p < 0.001)
Moukarzel, Di Rienzo, et al. (2019), Moukarzel, Guillot, et al. (2019)

EG: Motor imagery training + physical therapy

CG: physical therapy

 4; 3; 60 Function (TUG)

Pre > Post (p < 0.001)

EG > CG (p < 0.001)
Briones‐Cantero et al. (2020)

EG: Motor imagery training + physical therapy

CG: physical therapy

 NR; NR; 30 Pain (EVA); Function (WOMAC)

Pain:

Pre > Post (p < 0.001)

EG < CG (p < 0.001)

Function:

Pre < Post (p < 0.001)

EG > CG (p < 0.001)
Öztürk et al. (2021)

EG: Action observation training + physical therapy

CG: Physical therapy

 6;3;60 Pain (EVA); Function (WOMAC)

Pain:

Pre > Post (p < 0.001)

EG<CG (NS)

Function:

Pre < Post (p < 0.001)

EG > CG (NS)
Candiri et al. (2023)

EG: Motor imagery training + physical therapy

CG: Physical therapy

 6; 3; NR Pain (EVA); Function (TUG)

Pain:

Pre > Post (p < 0.001)

EG < CG (p = 0.008)

Function:

Pre < Post (p < 0.001)

EG > CG (p = 0.008)
Open in a new tab

Abbreviations: LPS, Latin Pain Scale; MI, motor imagery.

We applied the GRADE recommendations to evaluate the level of evidence for the use of movement representation strategies combined with standard physical therapy versus standard physical therapy in painful knee for pain and function. We observed a low recommendation for pain management due to the inconsistency highlighted by the high heterogeneity in the meta‐analyses (I 2 = 81%) and the imprecision of the data (n = 197). In the case of function, we observed a moderate recommendation due to the imprecision (n = 203; Figure 3).

FIGURE 3.

FIGURE 3

Open in a new tab

GRADE results.

3.4. Results of Meta‐Analyses

Results in pain outcomes of the application of graded motor imagery were shown in Figure 4. The pooled mean difference showed a significant overall effect of graded motor imagery intervention in favour of the experimental group compared to the control group (MD = −1.48; 95%; CI = −2.45, −0.50; p = 0.003). Results showed high heterogeneity and variability with an I 2 = 81%, not attributable to chance. Due to the high heterogeneity obtained in the analysis, we performed a sensitivity meta‐analysis excluding the study of Villafañe et al. (2017) due to its weight in the analysis. However, we found similar results with the exclusion of this study; a significant difference in favour of the experimental group remained (MD = −1.78, 95% CI = −2.80; −0.77; p = 0.0006).

FIGURE 4.

FIGURE 4

Open in a new tab

Forest plot pain.

Results in function outcomes of the application of graded motor imagery were shown in Figure 5. The pooled mean difference showed a significant overall effect of graded motor imagery intervention in favour of the experimental group compared to the control group (MD = −0.59; 95%; CI = −0.89, −0.28; p = 0.0001). Results showed low heterogeneity and variability with an I 2 = 36% not attributable to chance.

FIGURE 5.

FIGURE 5

Open in a new tab

Forest plot function.

4. Discussion

The aim of this systematic review and meta‐analysis was to evaluate the effectiveness of SRM on pain and function in patients with knee pain. Our results showed that the implementation of MRS added to standard physical therapy led to significant improvements in the pain and function of patients with knee pain.

The studies included in our review identified motor imagery, action observation and visualisation as the most commonly used MRS techniques in addition to standard physical therapy. A previous review found that these strategies were also the most commonly used in knee and hip arthroplasty rehabilitation using MRS (Riquelme‐Hernández et al. 2022) By integrating motor imagery, action observation and related imagery techniques, the interventions included in our review are grounded in neurophysiological evidence suggesting that these strategies engage overlapping premotor–parietal cortical motor networks and enhance corticospinal excitability (Hardwick et al. 2018; Binks et al. 2023) Although our review did not aim to evaluate these mechanisms directly, this underlying rationale supports the plausibility of the observed clinical effects on pain and function. Moreover, recent meta‐analyses demonstrate that the combined effect of these interventions is robust across varying treatment durations, supporting our decision to pool data by focusing on shared neurophysiological mechanisms rather than segregating by aetiology or intervention dosage (Chye et al. 2022).

The use of these strategies along with standard physical therapy treatment in our review showed a greater reduction in pain in patients compared to those who received standard physical therapy treatment alone. This is consistent with the previous review by Li et al. (2022) who examined the effects of the motor imagery strategy only in knee arthroplasty patients and found statistically significant effects on pain in these patients. The common standard physiotherapy alongside MRS in our included experimental groups mirrors common clinical practice and ensures that comparisons remain fair—controls received standard physiotherapy alone, while the experimental groups received the same physiotherapy base plus MRS. Previous reviews have similarly evaluated the additive effect of motor imagery or action observation on top of usual physiotherapy, with conclusions in the same line as our review (Cuenca‐Martínez et al. 2022; Zhao et al. 2023).

When interpreting our results across the included studies, the heterogeneity of outcome measures may limit the precision of our conclusions. However, our review is consistent with previous exploratory investigations that intentionally aggregated various functional and pain‐related metrics to generate preliminary evidence in a novel context where targeted research remains limited (Paravlic 2022; Li et al. 2022).

The review by Galonski et al. (2023) that evaluated the efficacy of motor imagery in patients with knee pain did not find statistically significant results in their study. This may be because they only considered this type of MRS and were only able to meta‐analyse the effects in patients with knee arthroplasty.

Our systematic review highlighted the positive effects of MRS, such as motor imagery and action observation, as adjuncts in rehabilitation for various knee conditions. These strategies appear to enhance pain reduction and functional improvements when combined with standard physical therapy. For instance, the randomised controlled trial by Tedeschi and collaborators in 2024 (Tedeschi et al. 2024) on the effectiveness of tele‐rehabilitation in patients with knee osteoarthritis complements our findings by demonstrating that remote rehabilitation interventions can successfully incorporate MRS to improve patient outcomes. This suggests that MRS, delivered via tele‐rehabilitation platforms, could be a feasible and effective option to expand access to rehabilitation services, particularly for populations facing barriers to in‐person care. Future research should explore the integration of MRS within tele‐rehabilitation frameworks to optimise treatment protocols for knee osteoarthritis and other knee pathologies.

When considering function, no significant improvements were observed across the studies when assessed by the time up and go test. In our review, statistically significant improvements in function were observed in favour of patients who received MRS in combination with standard physical therapy. In our study, in addition to the time up and go test, patient function was assessed using other tools such as the WOMAC questionnaire or the chair sit‐to‐stand test, and improvements were observed in favour of the experimental group regardless of the test used. Our results are consistent with the findings of the review by Cuenca‐Martínez et al. (2022) who observed the benefits of this type of strategy on functionality.

These results of functionality despite the presence of pain can be justified because the MRS strategies allowed physical and mental activation without the need for overt movement, as well as corticomotor excitability and modulation of intracortical inhibition (Lebon, Byblow, et al. 2012; Lebon, Guillot, et al. 2012; Paravlic 2022). This physiological support together with our results allows us to conclude the effectiveness of this type of strategy on pain and function in patients with knee pain.

4.1. Limitations

Our systematic review has several limitations that should be taken into account when interpreting the findings. Despite the results obtained from the ROB and GRADE assessments, the variation in the MRS and the meta‐analysis could have influenced the overall quality of the study. Most of the studies included were randomised controlled trials; however, some were pilot studies of such trials, so the results should be interpreted with caution. While the MRS showed positive effects on pain and function, future research should observe the effects in more pathologies that cause knee pain in order to analyse by subgroups and compare the effectiveness of such strategies. Lastly, although we conducted a comprehensive search of multiple electronic databases, including both publicly accessible and unpublished studies, some relevant studies may have been overlooked.

5. Conclusion

The results of our review suggest potential benefits from the implementation of MRS in addition to standard physical therapy for reducing pain and improving function in patients with knee pain. However, these findings must be interpreted with caution due to the considerable heterogeneity among the included studies in terms of population characteristics, intervention types, outcome measures and treatment duration. While the meta‐analyses revealed statistically significant improvements in both pain and function, the variability in study design and methodological quality limits the certainty and generalisability of the conclusions.

Considering these findings, future research should prioritise well‐designed randomised controlled trials that standardise outcome measures to improve comparability and enhance the robustness of our conclusions. Studies exploring optimal dosing, timing and combinations of movement representation strategies techniques with different rehabilitation protocols are needed to refine clinical guidelines. Additionally, research should investigate the long‐term effects of movement representation strategies techniques on pain and function, as well as patient adherence and acceptability in diverse knee pain populations. Clinically, incorporating movement representation strategies techniques as an adjunct to standard physical therapy may be recommended to enhance pain management and functional recovery, especially in patients with limited capacity for active movement. Practitioners should tailor these interventions considering individual patient profiles and therapy goals, while remaining aware of the current evidence base limitations.

Author Contributions

Conceptualisation and data collection were carried out by Julie Flebus, Javier Martín Núñez and María Granados Santiago; methodology and patient recruitment were carried out by Alejandro Heredia Ciuró and Andrés Calvache Mateo; analysis and writing of the article were carried out by Julie Flebus, Javier Martín Núñez and Alba Navas Otero with the assistance and supervision of Marie Carmen Valenza.

Ethics Statement

The authors have nothing to report.

Consent

The authors have nothing to report.

Conflicts of Interest

The authors declare no conflicts of interest.

Flebus, J. , Martín Núñez J., Granados Santiago M., et al. 2025. “Effectiveness of Movement Representation Strategies in Knee Pain: A Systematic Review and Meta‐Analyses on Knee Pain and Function.” European Journal of Pain 29, no. 9: e70100. 10.1002/ejp.70100.

Funding: Javier Martín Núñez and Alba Navas Otero received a grant supported by the Spanish Ministry of Education (Formación de Profesorado Universitario [FPU:21/00451 and FPU:22/01543]).

References

  1. Andrews, J. , Guyatt G., Oxman A. D., et al. 2013. “GRADE Guidelines: 14. Going From Evidence to Recommendations: The Significance and Presentation of Recommendations.” Journal of Clinical Epidemiology 66, no. 7: 719–725. 10.1016/j.jclinepi.2012.03.013. [DOI] [PubMed] [Google Scholar]
  2. Balshem, H. , Helfand M., Schünemann H. J., et al. 2011. “GRADE Guidelines: 3. Rating the Quality of Evidence.” Journal of Clinical Epidemiology 64, no. 4: 401–406. 10.1016/j.jclinepi.2010.07.015. [DOI] [PubMed] [Google Scholar]
  3. Bedson, J. , Mottram S., Thomas E., and Peat G.. 2007. “Knee Pain and Osteoarthritis in the General Population: What Influences Patients to Consult?” Family Practice 24, no. 5: 443–453. 10.1093/fampra/cmm036. [DOI] [PubMed] [Google Scholar]
  4. Bennell, K. L. , Hunter D. J., and Hinman R. S.. 2012. “Management of Osteoarthritis of the Knee.” BMJ 345: e4934. 10.1136/bmj.e4934. [DOI] [PubMed] [Google Scholar]
  5. Binks, J. A. , Emerson J. R., Scott M. W., Wilson C., van Schaik P., and Eaves D. L.. 2023. “Enhancing Upper‐Limb Neurorehabilitation in Chronic Stroke Survivors Using Combined Action Observation and Motor Imagery Therapy.” Frontiers in Neurology 14: 1097422. 10.3389/fneur.2023.1097422. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Briones‐Cantero, M. , Fernández‐de‐Las‐Peñas C., Lluch‐Girbés E., et al. 2020. “Effects of Adding Motor Imagery to Early Physical Therapy in Patients With Knee Osteoarthritis Who Had Received Total Knee Arthroplasty: A Randomized Clinical Trial.” Pain Medicine 21, no. 12: 3548–3555. 10.1093/pm/pnaa103. [DOI] [PubMed] [Google Scholar]
  7. Bunt, C. W. , Jonas C. E., and Chang J. G.. 2018. “Knee Pain in Adults and Adolescents: The Initial Evaluation.” American Family Physician 98, no. 9: 576–585. [PubMed] [Google Scholar]
  8. Candiri, B. , Talu B., Guner E., and Ozen M.. 2023. “The Effect of Graded Motor Imagery Training on Pain, Functional Performance, Motor Imagery Skills, and Kinesiophobia After Total Knee Arthroplasty: Randomized Controlled Trial.” Korean Journal of Pain 36, no. 3: 369–381. 10.3344/kjp.23020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Centre for Reviews and Dissemination . 2009. Systematic Reviews: CRD's Guidance for Undertaking Reviews in Health Care. University of York. https://www.york.ac.uk/crd/guidance/. [Google Scholar]
  10. Chye, S. , Valappil A. C., Wright D. J., et al. 2022. “The Effects of Combined Action Observation and Motor Imagery on Corticospinal Excitability and Movement Outcomes: Two Meta‐Analyses.” Neuroscience and Biobehavioral Reviews 143: 104911. 10.1016/j.neubiorev.2022.104911. [DOI] [PubMed] [Google Scholar]
  11. Cuenca‐Martínez, F. , Angulo‐Díaz‐Parreño S., Feijóo‐Rubio X., et al. 2022. “Motor Effects of Movement Representation Techniques and Cross‐Education: A Systematic Review and Meta‐Analysis.” European Journal of Physical and Rehabilitation Medicine 58, no. 1: 94–107. 10.23736/S1973-9087.21.06893-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. Cupal, D. D. , and Brewer B. W.. 2001. “Effects of Relaxation and Guided Imagery on Knee Strength, Reinjury Anxiety, and Pain Following Anterior Cruciate Ligament Reconstruction.” Rehabilitation Psychology 46, no. 1: 28–43. [Google Scholar]
  13. Fadiga, L. , Buccino G., Craighero L., Fogassi L., Gallese V., and Pavesi G.. 1999. “Corticospinal Excitability Is Specifically Modulated by Motor Imagery: A Magnetic Stimulation Study.” Neuropsychologia 37, no. 2: 147–158. 10.1016/s0028-3932(98)00089-x. [DOI] [PubMed] [Google Scholar]
  14. Fransen, M. , McConnell S., Harmer A. R., Van der Esch M., Simic M., and Bennell K. L.. 2015. “Exercise for Osteoarthritis of the Knee: A Cochrane Systematic Review.” British Journal of Sports Medicine 49, no. 24: 1554–1557. 10.1136/bjsports-2015-095424. [DOI] [PubMed] [Google Scholar]
  15. Galonski, T. , Mansfield C., Moeller J., Miller R., Rethman K., and Briggs M. S.. 2023. “Does Graded Motor Imagery Benefit Individuals With Knee Pain: A Systematic Review and Meta‐Analysis.” Journal of Bodywork and Movement Therapies 35: 130–139. 10.1016/j.jbmt.2023.05.005. [DOI] [PubMed] [Google Scholar]
  16. Grush, R. 2004. “The Emulation Theory of Representation: Motor Control, Imagery, and Perception.” Behavioral and Brain Sciences 27, no. 3: 377–442. 10.1017/s0140525x04000093. [DOI] [PubMed] [Google Scholar]
  17. Guyatt, G. H. , Oxman A. D., Vist G. E., et al. 2008. “GRADE: An Emerging Consensus on Rating Quality of Evidence and Strength of Recommendations.” BMJ 336, no. 7650: 924–926. 10.1136/bmj.39489.470347.AD. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Hardwick, R. M. , Caspers S., Eickhoff S. B., and Swinnen S. P.. 2018. “Neural Correlates of Action: Comparing Meta‐Analyses of Imagery, Observation, and Execution.” Neuroscience and Biobehavioral Reviews 94: 31–44. 10.1016/j.neubiorev.2018.08.003. [DOI] [PubMed] [Google Scholar]
  19. Higgins, J. P. T. 2011. Cochrane Handbook for Systematic Reviews of Interventions (Version 5.1.0). Cochrane Collaboration. https://training.cochrane.org/handbook/archive/v5.1/. [Google Scholar]
  20. Lebon, F. , Byblow W. D., Collet C., Guillot A., and Stinear C. M.. 2012. “The Modulation of Motor Cortex Excitability During Motor Imagery Depends on Imagery Quality.” European Journal of Neuroscience 35, no. 2: 323–331. 10.1111/j.1460-9568.2011.07938.x. [DOI] [PubMed] [Google Scholar]
  21. Lebon, F. , Guillot A., and Collet C.. 2012. “Increased Muscle Activation Following Motor Imagery During the Rehabilitation of the Anterior Cruciate Ligament.” Applied Psychophysiology and Biofeedback 37, no. 1: 45–51. 10.1007/s10484-011-9175-9. [DOI] [PubMed] [Google Scholar]
  22. Li, R. , Du J., Yang K., Wang X., and Wang W.. 2022. “Effectiveness of Motor Imagery for Improving Functional Performance After Total Knee Arthroplasty: A Systematic Review With Meta‐Analysis.” Journal of Orthopaedic Surgery and Research 17, no. 1: 65. 10.1186/s13018-022-02946-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Lotze, M. , and Halsband U.. 2006. “Motor Imagery.” Journal of Physiology, Paris 99, no. 4–6: 386–395. 10.1016/j.jphysparis.2006.03.012. [DOI] [PubMed] [Google Scholar]
  24. McAlindon, T. E. , Bannuru R. R., Sullivan M. C., et al. 2014. “OARSI Guidelines for the Non‐Surgical Management of Knee Osteoarthritis.” Osteoarthritis and Cartilage 22, no. 3: 363–388. 10.1016/j.joca.2014.01.003. [DOI] [PubMed] [Google Scholar]
  25. Moukarzel, M. , Di Rienzo F., Lahoud J. C., et al. 2019. “The Therapeutic Role of Motor Imagery During the Acute Phase After Total Knee Arthroplasty: A Pilot Study.” Disability and Rehabilitation 41, no. 8: 926–933. 10.1080/09638288.2017.1419289. [DOI] [PubMed] [Google Scholar]
  26. Moukarzel, M. , Guillot A., Di Rienzo F., and Hoyek N.. 2019. “The Therapeutic Role of Motor Imagery During the Chronic Phase After Total Knee Arthroplasty: A Pilot Randomized Controlled Trial.” European Journal of Physical and Rehabilitation Medicine 55, no. 6: 806–815. 10.23736/S1973-9087.19.05136-0. [DOI] [PubMed] [Google Scholar]
  27. Neogi, T. , Nevitt M. C., Yang M., Curtis J. R., Torner J., and Felson D. T.. 2010. “Consistency of Knee Pain: Correlates and Association With Function.” Osteoarthritis and Cartilage 18, no. 10: 1250–1255. 10.1016/j.joca.2010.08.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Öztürk, Ö. , Bombacı H., Keçeci T., and Algun Z. C.. 2021. “Effects of Additional Action Observation to an Exercise Program in Patients With Chronic Pain due to Knee Osteoarthritis: A Randomized‐Controlled Trial.” Musculoskeletal Science & Practice 52: 102334. 10.1016/j.msksp.2021.102334. [DOI] [PubMed] [Google Scholar]
  29. Page, S. J. , Levine P., and Hill V.. 2015. “Mental Practice – Triggered Electrical Stimulation in Chronic, Moderate, Upper‐Extremity Hemiparesis After Stroke.” American Journal of Occupational Therapy 69, no. 1: 6901290050p1–6901290050p8. 10.5014/ajot.2015.014902. [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Paravlic, A. H. 2022. “Motor Imagery and Action Observation as Appropriate Strategies for Home‐Based Rehabilitation: A Mini‐Review Focusing on Improving Physical Function in Orthopedic Patients.” Frontiers in Psychology 13: 826476. 10.3389/fpsyg.2022.826476. [DOI] [PMC free article] [PubMed] [Google Scholar]
  31. Paravlic, A. H. , Pisot R., and Marusic U.. 2019. “Specific and General Adaptations Following Motor Imagery Practice Focused on Muscle Strength in Total Knee Arthroplasty Rehabilitation: A Randomized Controlled Trial.” PLoS One 14, no. 8: e0221089. 10.1371/journal.pone.0221089. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Park, S. D. , Song H. S., and Kim J. Y.. 2014. “The Effect of Action Observation Training on Knee Joint Function and Gait Ability in Total Knee Replacement Patients.” Journal of Exercise Rehabilitation 10, no. 3: 168–171. 10.12965/jer.140112. [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Peat, G. , McCarney R., and Croft P.. 2001. “Knee Pain and Osteoarthritis in Older Adults: A Review of Community Burden and Current Use of Primary Health Care.” Annals of the Rheumatic Diseases 60, no. 2: 91–97. 10.1136/ard.60.2.91. [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Riquelme‐Hernández, C. , Reyes‐Barría J. P., Vargas A., et al. 2022. “Effects of the Practice of Movement Representation Techniques in People Undergoing Knee and Hip Arthroplasty: A Systematic Review.” Sports 10, no. 12: 198. 10.3390/sports10120198. [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Schuster, C. , Hilfiker R., Amft O., et al. 2011. “Best Practice for Motor Imagery: A Systematic Literature Review on Motor Imagery Training Elements in Five Different Disciplines.” BMC Medicine 9: 75. 10.1186/1741-7015-9-75. [DOI] [PMC free article] [PubMed] [Google Scholar]
  36. Skou, S. T. , Roos E. M., Laursen M. B., et al. 2015. “A Randomized, Controlled Trial of Total Knee Replacement.” New England Journal of Medicine 373, no. 17: 1597–1606. 10.1056/NEJMoa1505467. [DOI] [PubMed] [Google Scholar]
  37. Tedeschi, R. , Platano D., Pillastrini P., Berti L., and Benedetti M. G.. 2024. “Effectiveness of Tele‐Rehabilitation in Patients With Knee Osteoarthritis: A Randomized Controlled Trial.” Digital Health 10: 20552076241286186. 10.1177/20552076241286186. [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Thieme, H. , Morkisch N., Rietz C., Dohle C., and Borgetto B.. 2016. “The Efficacy of Movement Representation Techniques for Treatment of Limb Pain – A Systematic Review and Meta‐Analysis.” Journal of Pain 17, no. 2: 167–180. 10.1016/j.jpain.2015.10.015. [DOI] [PubMed] [Google Scholar]
  39. Turkiewicz, A. , Gerhardsson de Verdier M., Engström G., et al. 2015. “Prevalence of Knee Pain and Knee OA in Southern Sweden and the Proportion That Seeks Medical Care.” Rheumatology 54, no. 5: 827–835. 10.1093/rheumatology/keu409. [DOI] [PubMed] [Google Scholar]
  40. Villafañe, J. H. , Isgrò M., Borsatti M., Berjano P., Pirali C., and Negrini S.. 2017. “Effects of Action Observation Treatment in Recovery After Total Knee Replacement: A Prospective Clinical Trial.” Clinical Rehabilitation 31, no. 3: 361–368. 10.1177/0269215516642605. [DOI] [PubMed] [Google Scholar]
  41. Vlaeyen, J. W. S. , Crombez G., and Linton S. J.. 2016. “The Fear‐Avoidance Model of Pain.” Pain 157, no. 8: 1588–1589. 10.1097/j.pain.0000000000000574. [DOI] [PubMed] [Google Scholar]
  42. Wilczynska, D. , Lysak A., and Podczarska‐Glowacka M.. 2015. “Imagery Use in Rehabilitation After the Knee Joint Arthroscopy.” Baltic Journal of Health and Physical Activity 7, no. 4: 9. [Google Scholar]
  43. Wylde, V. , Beswick A., Bruce J., Blom A., Howells N., and Gooberman‐Hill R.. 2018. “Chronic Pain After Total Knee Arthroplasty.” EFORT Open Reviews 3, no. 8: 461–470. 10.1302/2058-5241.3.180004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  44. Zhao, L. J. , Jiang L. H., Zhang H., et al. 2023. “Effects of Motor Imagery Training for Lower Limb Dysfunction in Patients With Stroke: A Systematic Review and Meta‐Analysis of Randomized Controlled Trials.” American Journal of Physical Medicine & Rehabilitation 102, no. 5: 409–418. 10.1097/PHM.0000000000002107. [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from European Journal of Pain (London, England) are provided here courtesy of Wiley

RESOURCES