Total knee arthroplasty (TKA) is accepted as the gold standard treatment in the last phase of knee osteoarthritis [1]. It is one of the most common surgeries worldwide and demand is expected to quadruple by 2030 [2]. Although surgery provides relief of many symptoms, some symptoms may persist. Commonly reported symptoms are pain, stiffness, joint sounds, edema, and difficulty in the activities of daily living [3]. Chronic pain, which is one of the biggest causes of postoperative dissatisfaction, has been reported in 15%–20% of individuals with TKA [4,5]. While the cause of the pain may be post-surgical problems, in some cases there is chronic persistent pain without any problems after surgery. The presence of long-term osteoarthritis and central sensitization before surgery is important. In addition, postoperative pain and immobilization are other factors [5–7]. Postoperative intensive rehabilitation provides improvement of muscle strength and function. However, patients’ results cannot return to a condition equal to that of healthy individuals. Therefore, complementary therapeutic approaches are important to increase the effect of rehabilitation in the acute period [8]. In addition, it is a common view that in chronic conditions involving the musculoskeletal system, focusing only on structural changes in the periphery is not sufficient. It is emphasized that central changes should be targeted by taking into account the rapid adaptation of the central nervous system [9]. For this purpose, movement representation techniques including mirror therapy, action observation therapy, graded motor imagery (GMI), and motor imagery can be used in rehabilitation [10]. Movement representation techniques are an effective treatment option for musculoskeletal problems in terms of pain, muscle strength, mobility, and function [7].

The background of GMI is based on neuromatrix and neuroplasticity mechanisms [11]. In chronic pain, minimal nociceptive and non-nociceptive inputs cause pain generation, thus strengthening the pain neuromatrix. The neuromatrix approach focuses on the controlled activation of components of the pain neuromatrix and aims to increase tolerance to irritating inputs [12]. However, chronic musculoskeletal disorders cause maladaptive plastic reorganization in the somatosensory and motor cortex, both structurally and functionally [9]. GMI emphasizes neuroplasticity through restoring altered cortical body maps, gradual activation of cortical activation, and reduction of cortical disinhibition [13,14]. GMI consists of right-left discrimination (lateralization), motor imagery, and mirror therapy stages [15]. As far as the authors know, no studies on GMI after TKA have been found. However, motor imagery, which is one of the stages of GMI, caused significant changes in pain and disability related to pain, edema, range of motion (ROM), and strength in postoperative acute and chronic rehabilitation [7,8]. Motor imagery showed beneficial effects on walking performance and functional recovery 6 months after surgery [16]. Orthopedic patients without any cognitive problems can benefit from motor imagery-based mental exercise techniques at the maximum level [6]. GMI added to the rehabilitation program in individuals who have undergone lumbar surgery and fracture surgery has shown significant contributions to recovery [15,17–19]. Therefore, this study was designed to investigate the effect of the GMI program added to rehabilitation in the early period after TKA surgery on pain, functional performance, motor imagery skills, and kinesiophobia.

1. Trial design

This study was designed as a prospective, randomized controlled study investigating GMI in patients with TKA. The study was approved by Inonu University Clinical Research Ethics Committee (approval number: 2021/173) and registered in the Clinical Trial (registration number: NCT05138406). This research was carried out between October 2021 and December 2022 and conducted in accordance with the Declaration of Helsinki. Informed consent was obtained from all participants.

2. Participants

Individuals between the ages of 45 and 80 who were diagnosed with knee osteoarthritis and planning unilateral TKA were included in the study. After the patients were operated on with the same surgical technique (anterior approach, cemented), they had to be suited to early postoperative physiotherapy by the orthopedist. Exclusion criteria included revision/bilateral TKA, contralateral knee osteoarthritis (greater than 4/10 pain with activity), any physical therapy intervention or any other knee surgery in the past 6 months, any neurological, cardiac, pulmonary, surgery, or psychiatric illness in pre- or post-surgery periods, acute postoperative infection or fever, cognitive impairments that altered the probability of correct understanding of the motor imagery program, body mass index > 35 kg/m², comprehension or communication difficulties, and insufficient knowledge of Turkish to follow study instructions.

3. Sample size

Power analysis was performed with α = 0.05 and 1-β (power) = 0.80 using the public statistics software Openepi version 3 (http://www.openepi.com) to determine the number of samples to be included in the study. The calculation was made by assuming that the difference between Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) pain division scores before rehabilitation (11.05 ± 3.02) and post-rehabilitation (7.16 ± 2.67) in individuals with TKA was 3.89 units [20]. It was determined that a total of 18 people, including at least 9 people in each group should be included.

4. Randomization

Patients were selected from the relevant population using the non-probability random sampling method (the first random patient encountered). Individuals who agreed to participate in the study and met the inclusion criteria were assigned to groups (GMI or control) using numbered (in order and equal numbers for the two groups) closed envelopes.

5. Interventions

Participants were included in the rehabilitation program 3 times a week for 6 weeks in both the GMI group and control group (CG). The treatment program was started 24–48 hours after surgery. The GMI group received traditional rehabilitation similar to CG in addition to GMI. In addition, the same pain control method was applied to all participants. Paracetamol was administered intravenously (4 × 1) on the 1st postoperative day. The dose was gradually reduced according to the patient's pain threshold during the hospitalization period, which was 2–3 days on average. After discharge, recommendations were made by the orthopedist regarding the use of paracetamol.

6. GMI

The GMI program was administered by a certified physiotherapist. The stages were lateralization, motor imagery, and mirror therapy, consecutively. Each stage was implemented for 2 weeks. The program was in the form of a physiotherapist (supervised) 3 days a week and a home program 3 days a week. In line with the research of Lagueux et al. [21], each stage was applied 3 times a day for 10 minutes. In the lateralization phase, 60 images obtained from the Recognise™ Knee application were used. In order to increase the difficulty of this application, Basic, Vanilla, and Context stages were used respectively. When any stage was completed comfortably, the next stage was undertaken. Continuity of the practice outside of the supervised sessions was encouraged by the participants. For this purpose, the pictures were reproduced and delivered to the participants during the first session. Participants were asked to distinguish right/left knee images without focusing on their knees. During the motor imagery phase, participants were asked to focus fully on the imagery of movements in a quiet environment. Two methods were used for the imagined movements. In order to visualize the movements for the normal movement of the knee, the images used in the lateralization phase, especially those showing the maximum flexion of the knee, were preferred. In addition, participants were asked to imagine different activities in daily life. These progressed from the visualization of simple to more difficult activities. Sitting on the toilet, getting up from the toilet, sitting on a chair, going up and down stairs, cycling, going uphill, downhill, standing on one leg, walking on different surfaces (grass, stony road, sand), going to the sidewalk, getting off the sidewalk, washing dishes standing up, as well as driving and running activities were imagined. In the mirror therapy phase, a specially designed mirror box and a portable mirror were used. For exercises in the supine position, the operated extremity was placed in the mirror box. While the participant was watching the reflection of the healthy limb in the mirror, they were asked to do some exercises. These included ankle pumping, hip and knee flexion, straight leg raise, as well as hip abduction and adduction. In addition, a portable mirror was used for exercises in sitting position. Knee flexion and extension exercises were performed (Fig. 1).

Figure 1. Graded motor imagery program.

7. Traditional rehabilitation

The first-week program consisted of ankle-pumping exercises, passive/active/active assisted hip flexion (straight leg raise), abduction/adduction, passive/active/active assisted hip-knee flexion (starting from 30°–40°), isometric exercises (quadriceps, adductor, gluteal muscles, hamstrings), cryotherapy (3 times a day), and stretching exercises (hamstring, gastro-soleus). In addition, strengthening exercises for upper extremity and trunk muscles, sitting by the bed, knee flexion extension in sitting, standing up with the help of a walker, walking training, and ladder training was performed [7]. The exercises in the first week had been continued between 2–6 weeks. In addition, strengthening exercises (all exercises in the first stage with the help of weights/theraband), training for the stages of walking (heel strike, etc.), stepping to the side, standing on one leg, and advanced functional activities were performed [8,22].

8. Outcome measures

Age, sex, body weight, height, body mass index (kg/m²), and the disease-related information (affected side, duration of pain) of the included patients were recorded. The adherence of the participants to the sessions was also calculated as a percentage. All measurements were performed by an experienced physiotherapist the day before surgery and after 6 weeks of rehabilitation.

9. Primary outcomes

Activity and rest pain intensity in the knee joint were evaluated with the help of the visual analogue scale. A horizontal line 10 cm in length was used, with endpoints indicating “0: No pain” and “10: Unbearable pain”. Participants were asked to mark the severity of pain on this line. Pain intensity was measured in cm [23].

The WOMAC was used for functional assessment. WOMAC consists of pain, stiffness, and physical function subsections. Higher scores indicate more pain, stiffness, and worse physical function [24].

10. Secondary outcomes

A standard goniometer was used to assess the active ROM. For knee flexion ROM, the fixed arm of the goniometer was parallel to the long axis of the femur. The movable arm was placed parallel to the long axis of the fibula. The participant was asked to flex the knee without disturbing the contact of the foot with the bed. The extension angle was measured with the same procedure, after the heel was placed at a height of approximately 15 cm, and after active knee extension was requested from the participant. Measurements for flexion and extension were repeated 3 times and their averages were recorded [7,25] (Fig. 2A, B).

Figure 2. (A) Knee flexion range of motion measurement. (B) Knee extension range of motion measurement. (C) Hand-held dynamometer. (D) Quadriceps femoris muscle force measurement.

Muscle strength was measured with the help of a hand-held dynamometer (Lafayette Instrument_®) with the knee in a 90° flexion position at the bedside. The device was placed on the anterior face above the malleoli in the ankle. Maximum isometric contraction in the extension direction was requested for 5 seconds against the device. Three repetitions of the movement were performed and averaged. A 30-second rest break was given between each contraction. Measurements were recorded in kilograms [26] (Fig. 2C, D).

Functional mobility was assessed using the Timed Up and Go Test (TUG). A chair, 46 cm in height, was used to perform the test and the distance of 3 m from this chair was marked. The participants were asked to sit on the chair and go 3 m as fast as possible without any support from their arms, and then turn around and sit on the chair. The total time for this task was measured with a stopwatch. The time was recorded in hundredths of a second [27] (Fig. 3A).

Figure 3. (A) Measurement of timed up and go test. (B) Measurement of mental chronometer performance.

Explicit and implicit motor imagery abilities were evaluated separately. A mental chronometer and the Movement Imagery Questionnaire-3 (MIQ-3) were used to assess explicit motor imagery ability. The laterality task was used for implicit motor imagery [28].

Mental chronometer performance was measured by the imaging of the TUG. When the participant completed the visualization task and sat on the chair, the participant was asked to say “stop” and the time was stopped [6]. A negative mental chronometer time indicates that the imagined TUG (iTUG) performance is longer than the actual performance; being positive indicates that the iTUG performance is shorter. In this task, the following formula was used to calculate the mental chronometer time [29] (Fig. 3B).

The MIQ-3 included internal and external visual imagery and kinesthetic imagery of four different movements. First of all, the physical movement in the questionnaire was performed, and then the imagery stages were carried out for the same movement. Participants rated the clarity of the movement they imagined on an ordinal scale with 7 points (1 = Very hard to feel/see, 7 = Very easy to feel/see). The score of each section was calculated separately, and then the scores of the sections were summed and the total score was recorded. Higher scores indicate better mental imagery ability [30].

For knee lateralization performance, the “Vanilla” part of the Recognise™ Knee application developed by the Neuro Orthopedic Institute was used. A total of 20 knee-related pictures were shown for 5 seconds. Participants were asked to press the right or left button on the screen as accurately and quickly as possible without focusing on the knee. Participants were allowed to experiment before the actual implementation. Accuracy rates and the reaction time were recorded [31].

The Pain Catastrophizing Scale (PCS) was used to evaluate feelings and thoughts about pain. The PCS consists of 13 items in total. Each question is scored on a 5-point scale (0: Never, 1: Somewhat, 2: Moderately, 3: Seriously, 4: Always). The scale consists of 3 sub-sections, namely, helplessness, magnification, and rumination. High scores indicate high pain catastrophizing [32,33].

Part A of the Central Sensitization Inventory (CSI) was used for central sensitization. Twenty-five questions evaluating symptoms related to central sensitization are scored on a 5-point Likert scale from 0 to 4 (0 = Never, 4 = Always). A score above 40 indicates the presence of central sensitization and higher scores indicate higher central sensitization symptoms [34,35].

The Tampa Kinesiophobia Scale (TKS) was used to assess fear of movement (kinesiophobia). The scale uses a 4-point Likert scoring system (1 = Strongly disagree, 2 = Disagree, 3 = Agree, 4 = Strongly agree). Higher scores indicate higher kinesiophobia [36].

11. Statistical analysis

Basic demographic and clinical variables between groups were analyzed with the Mann–Whitney U-test (for continuous variables) and the chi-square test (for categorical variables). The Mann–Whitney U-test was used to compare the pre- and post-treatment outcome measures between the groups. In addition, the comparison of pre- and post-treatment outcome measures within the group was analyzed with the Wilcoxon signed-rank test. The Cohen’s d was calculated for the effect of the interventions. Cohen’s d was interpreted as insignificant (< 0.2), small (0.2–0.5), moderate (0.5–0.8), or large (> 0.8) [37]. P < 0.05 was considered statistically significant. Statistical analysis was done using Software SPSS 25.0 version (IBM Co.).

In total, 38 patients were screened for eligibility. Twenty individuals met the eligibility criteria and agreed to participate in the study. Ten of the participants were randomized to the GMI group and 10 to the GC. Two individuals discontinued the study, and as a result, 9 individuals in the groups were included in the analysis. The details are in Fig. 4. The adherence of the participants to the sessions in both groups was 94.44. A comparison of the basic demographic and clinical characteristics of the participants is presented in Table 1. Throughout the study, participants did not report any adverse effects. All pre-treatment clinical measurements of the participants were similar (Tables 2, 3).

Table 1

Participants’ baseline demographic and clinical characteristics

Characteristic		GMI (n = 9)	Control (n = 9)	P value
Age (yr)		64.77 ± 4.52	67.22 ± 6.61	0.142a
Body mass index (kg/m2)		31.30 ± 3.19	29.70 ± 3.08	0.289a
Pre-operative pain (yr)		11.55 ± 6.42	11.00 ± 4.55	0.892a
Sex	Female	8 (88.9)	8 (88.9)	> 0.999b
	Male	1 (11.1)	1 (11.1)
Injury side	Right	5 (55.6)	5 (55.6)	> 0.999b
	Left	4 (44.4)	4 (44.4)

Values are presented as the mean ± standard deviation or number (%).

GMI: graded motor imagery.

Significance was accepted as P < 0.05. ^aMann–Whitney U-test, ^bChi-squared test.

Table 2

Comparison of pain intensity and functional assessments of participants in the GMI and control groups

Variable		Baseline	P valuea	Post treatment	P valuea	P valueb	Cohen’s d
Pain intensity, cm
VAS-activity	GMI (n = 9)	7.00 (6.55–9.00)	0.825	1.90 (1.25–2.45)	< 0.001	0.008	5.12
	Control (n = 9)	8.40 (6.55–8.95)		4.30 (3.70–4.70)		0.008	3.99
VAS-rest	GMI (n = 9)	5.70 (3.70–7.00)	0.757	1.00 (0.25–1.40)	0.004	0.008	2.63
	Control (n = 9)	5.50 (1.50–7.05)		2.60 (2.00–3.05)		0.092	0.84
WOMAC	GMI (n = 9)	21.09 (16.46–26.79)	0.102	6.79 (5.20–8.88)	0.145	0.008	3.06
	Control (n = 9)	26.87 (24.64–28.99)		8.08 (7.16–10.55)		0.008	3.39

Values are presented as median (interquartile range).

GMI: graded motor imagery, VAS: visual analog scale, WOMAC: Western Ontario and McMaster Universities Osteoarthritis Index.

^aMann–Whitney U-test, ^bWilcoxon signed-rank test. Significance was accepted as P < 0.05.

Table 3

Comparison of pre- and post-treatment secondary outcome measures of participants in the groups

Variable		Baseline	P valuea	Post treatment	P valuea	P valueb	Cohen’s d
Flexion (°)	GMI	104.33 (91.16–112.5)	0.427	105.00 (99.5–114.33)	0.200	0.021	0.47
	Control	94.66 (90.16–107.66)		102.33 (90.50–109.66)		0.079	0.38
Extension (°)	GMI	2.33 (2.00–3.00)	0.168	1.00 (1.00–1.83)	0.347	0.007	2.00
	Control	4.00 (1.00–6.83)		2.00 (0.00–3.66)		0.018	0.79
Quadriceps femoris (force, kg)	GMI	5.80 (4.86–7.03)	0.627	6.83 (5.73–7.96)	0.216	0.028	0.75
	Control	6.40 (4.50–8.26)		8.03 (6.90–8.23)		0.109	0.76
TUG (sec)	GMI	15.39 (11.40–16.89)	0.310	11.05 (9.44–13.11)	0.402	0.011	1.01
	Control	17.60 (12.33–21.22)		12.49 (9.99–15.44)		0.028	0.74
iTUG (sec)	GMI	6.07 (3.94–7.89)	0.085	4.60 (3.21–6.05)	0.085	0.008	0.81
	Control	8.32 (6.52–13.22)		5.30 (4.41–9.07)		0.066	0.79
Mental chronometry (%)	GMI	78.82 (51.36–103.07)	0.233	80.16 (74.18–107.27)	0.566	0.374	0.29
	Control	50.56 (26.30–84.29)		80.16 (54.49–84.62)		0.314	0.37
MIQ-3	GMI	51.00 (46.50–61.50)	0.331	64.00 (63.00–72.50)	0.037	0.008	1.92
	Control	47.00 (41.50–54.00)		59.00 (53.00–66.00)		0.008	1.49
Lateralization, Knee, affected
Recognition accuracy (%)	GMI	60.00 (45.00–70.00)	0.964	90.00 (80.00–90.00)	0.015	0.007	1.95
	Control	70.00 (40.00–75.00)		60.00 (55.00–80.00)		0.107	0.35
Response time (sec)	GMI	2.60 (2.50–3.60)	0.121	2.30 (1.80–2.45)	0.199	0.008	1.39
	Control	2.30 (2.00–3.00)		2.60 (2.05–4.15)		0.611	0.25
CSI-A	GMI	44.00 (31.50–46.00)	0.724	23.00 (11.50–26.00)	0.894	0.008	1.80
	Control	37.00 (13.50–52.50)		23.00 (6.00–28.50)		0.008	0.91
PCS	GMI	30.00 (19.00–34.00)	0.894	2.00 (0.00–4.00)	0.039	0.008	2.83
	Control	26.00 (16.00–34.50)		13.00 (2.00–19.00)		0.011	1.17
TKS	GMI	41.00 (35.50–42.50)	0.084	27.00 (26.50–31.00)	0.009	0.008	2.11
	Control	44.00 (39.50–54.00)		35.00 (30.50–39.50)		0.008	1.38

Values are presented as median (interquartile range).

GMI: graded motor imagery, TUG: Timed Up and Go Test, iTUG: imagined Timed Up and Go Test, MIQ-3: Movement Imagery Questionnaire-3, CSI-A: Central Sensitisation Inventory-A, PCS: Pain Catastrophizing Scale, TKS: Tampa Kinesiophobia Scale.

^aMann–Whitney U-test, ^bWilcoxon signed-rank test. Significance was accepted as P < 0.05.

Figure 4. Design of the study (CONSORT flow diagram).

1. Primary outcomes

Table 2 shows the comparison of primary outcome measures before and after treatment. Within the group, activity pain was significantly reduced in the GMI and GCs (P = 0.008). GMI and traditional rehabilitation showed strong effect sizes (Cohen’s d = 5.12 and d = 3.99, respectively). However, the reduction in the GMI group was significantly greater (P < 0.001). Resting pain decreased significantly only in the GMI group (P = 0.008; d = 2.63). In addition, there was a significant decrease in favor of the GMI group between the groups (P = 0.004). WOMAC scores differed significantly in both treatment groups (P = 0.008). There was no significant difference between the groups (P = 0.145).

2. Secondary outcomes

Table 3 shows the comparison of secondary outcome measures before and after treatment. The flexion ROM increased significantly only in the GMI group (P = 0.021; d = 0.47). However, there was no significant difference between the groups after treatment (P = 0.200). There were significant reductions in GMI and CG in extension ROM within the group (P = 0.007 and P = 0.018, respectively), but GMI showed a strong effect size (d = 2.00). The extension ROM did not differ between the groups (P = 0.347). The quadriceps femoris muscle strength was also not different between the groups (P = 0.216); A significant, moderate increase was observed in the GMI group within the group (P = 0.028; d = 0.75). TUG and iTUG were not significantly different between groups (P = 0.402 and P = 0.085, respectively) and within the group, a significant decrease was found in both the GMI and GCs. GMI showed a strong effect for the TUG and iTUG (d = 1.01 and d = 0.81, respectively). The mental chronometer was not significantly different within and between the groups. MIQ-3 resulted in significant improvement in favor of GMI between the groups (P = 0.037). In intragroup comparison, significant improvement was found in both the GMI and CG (P = 0.008, d = 1.92 and d = 1.49, respectively). The accuracy of the affected knee in knee lateralization performance improved significantly between the groups in favor of the GMI group (P = 0.015). Within the group, there was a significant and strong increase only in the GMI group (P = 0.007; d = 1.95). While the reaction time was not significantly different between the groups, a significant and strong decrease was shown in the GMI group (P = 0.008; d = 1.39). While central sensitization was not different between the groups after treatment, it resulted in a significant and strong reductions in intragroup GMI and CG (P = 0.008; respectively d = 1.80 and d = 0.91). There was a significant decrease in favor of the GMI group between the groups in PCS and TKS (P = 0.039 and P = 0.009, respectively). There were significant reductions in GMI and CG within the group.

This study was designed to investigate the effects of adding GMI to the acute rehabilitation program after TKA surgery, primarily on pain and functional status. In addition, its effects on a wide range of muscle strength, ROM, mobility, motor imagery ability, kinesiophobia and pain, as well as other parameters were examined in detail. GMI showed significant effects on pain, motor imagery performance, pain catastrophizing, and kinesiophobia compared to conventional rehabilitation.

The importance of targeting central sensory-motor integration and peripheral and sympathetic mechanisms for the activation of neural pathways is emphasized. For this purpose, motion representation techniques are used [38]. In a systematic review, significant effects of motion representation techniques on motor imagery ability, speed, pain, strength, and gait were reported in patients with total knee and hip arthroplasty [10]. In a study, structural and functional changes were observed in brain gray matter volume and nociceptive processing areas before and after TKA surgery. However, the presence of chronic pain in some cases after surgery suggests inconsistency with adaptations in the cortex [39]. Mentally-based exercises provide cortical motor activation without creating a protective pain response [11]. Motor imagery exercises applied in the acute period of TKA provided pain relief [7,8]. After lumbar spine surgery was applied in the management of pain and degenerative disorders, treatment programs including the stages of GMI were applied. Significant improvement in pain, kinesiophobia, disability, and ROM had been demonstrated after treatment [17,18]. The authors’ findings showed significant improvements in activity and rest pain. Interestingly, the WOMAC score improved significantly in both treatment groups and there was no difference between the groups. This may be related to the fact that this questionnaire actually has sub-dimensions such as pain, stiffness, and physical function [24]. In addition, the contribution of TKA surgery to the natural healing process may explain the lack of difference between the groups [7]. While GMI produced significant improvements in ROM, results were not different between the groups. The effects of motion representation techniques on ROM are contradictory. While it has been shown in a few studies that motor imagery does not cause significant improvements in ROM in TKA rehabilitation [7,40], another study showed positive effects in flexion ROM [8]. Randomized controlled studies have shown that GMI improves ROM in wrist, shoulder, and elbow pathologies [15,19,41]. The results of a systematic review stated that the limited effect of motor imagery on ROM in the acute period may be related to biomechanical changes such as fear of movement or edema of the tissues around the joint [42]. Since the authors’ results led to significant improvements in kinesiophobia, factors such as periarticular edema may have had a limiting effect.

Although GMI caused a significant increase in strength, there was no difference between the groups after treatment. In addition, the treatments showed similar effect sizes. There are limited and conflicting results regarding the effects of GMI on muscle strength after orthopedic surgeries. It produced effective results compared to traditional rehabilitation after elbow fracture surgery [19]. However, there was no change after distal radius fracture surgery [15]. In contrast, motor imagery in TKA caused significant improvement in muscle strength [8,16]. GMI is thought to increase muscle strength by increasing corticospinal excitability. However, post-surgery edema, pain, and factors during surgery cause significant muscle inhibition [40]. This could change the results of GMI. In addition, the focus on function and joint ROM during the stages of GMI may also have affected the results.

TUG test performance resulted in significant improvement in both groups. Although the results were not different between the groups, the GMI showed a strong effect (Cohen’s d = 2.88). The effects of motor imagery on TUG performance are also heterogeneous in the literature. Motor imagery added to the rehabilitation after acute surgery resulted in a significant reduction in TUG performance [6]. In another study, similar to the authors’ findings, although there were significant differences in the GMI and GCs, there was no significant difference between the groups [8]. TUG performance was associated with lower extremity muscle strength [43]. Therefore, other gains may affect performance. In the present study, the limited improvement in strength in the GMI group compared to the GC may have limited the TUG performance.

The accuracy rates on the lateralization performance (implicit motor imagery) also improved significantly. Motor imagery ability is considered a backdoor to facilitate movement [44]. In one study, motor imagery exercises caused a significant improvement in the motor imagery ability of patients with TKA [45]. In fact, improvement was expected in only the GMI group, while improvement in both treatment groups was interesting. However, it has been shown that physically active individuals have better motor imagery vividness [46]. This can be explained by the effect of the rehabilitation received by the GC.

It was interesting that the mental chronometer did not improve. True TUG and iTUG improved significantly after treatment. This was similar to the work of Zapparoli et al. [6]. But mental chronometer times also showed improvement [6]. This may be related to a further decrease in the iTUG in the present study. There is limited research on the effect of GMI on mental chronometer times. Future research is needed to better interpret the authors’ results. Lateralization performance is an indicator of the body schema. Chronic pain conditions of the musculoskeletal system cause deterioration in lateralization performance [47]. There was no improvement in lateralization performance in the group that received conventional rehabilitation after TKA surgery. However, significant improvement was shown in the GMI group. This suggests that it is necessary to add GMI to traditional rehabilitation. The findings of this study support the results of the study in which GMI was applied after elbow stiffness [19].

Central sensitization is an important factor in causing chronic pain related to the sensitivity of the nervous system [48]. Although CSI decreased significantly in both groups, it showed a stronger effect in the GMI group. Decreased sensitivity also contributes to the improvement of emotional and affective dimensions of pain, such as kinesiophobia and pain catastrophizing [49,50]. Parallel to this, pain catastrophizing and kinesiophobia showed significant improvement in favor of GMI. The effects of GMI after musculoskeletal problems (elbow contracture, shoulder problems) support the authors’ results [19,41,50].

This study has some limitations. First of all, the small sample size complicates a better interpretation of the results. The inclusion criteria, which we elaborated to better understand the effects of treatment, prevented us from reaching sufficient sampling within the planned timeframe. Another limitation is the authors’ focus on the acute effect. Therefore, long-term follow-up studies with more individuals will yield important results.

The strongest aspect of this research is its comprehensive assessment of the effect of treatment. The effects of GMI on pain, the affective dimensions of pain, and the problems caused by chronic pain have been studied in detail. It is also the first study to investigate the effect of GMI after TKA surgery.

GMI added to acute period rehabilitation after TKA led to significant improvements in pain severity, explicit and implicit motor imagery ability, pain catastrophizing, and kinesiophobia. However, no significant changes were observed in functional parameters. GMI is an effective and safe therapeutic tool that can be used to increase the effectiveness of rehabilitation from the acute period.

We would like to thank Associate professor Emre ERGEN for their patient referral contributions.

Data files are available from Harvard Dataverse: https://doi.org/10.7910/DVN/T4VB64. Further inquiries can be directed to the corresponding author.

No potential conflict of interest relevant to this article was reported.

This research was supported by Inonu University Scientific Research Projects Unit with Project number 2021/2743.

Busra Candiri: Writing/manuscript preparation; Burcu Talu: Writing/manuscript preparation; Emre Guner: Writing/manuscript preparation; Metehan Ozen: Writing/manuscript preparation.

1. Zeni JA Jr, Axe MJ, Snyder-Mackler L. Clinical predictors of elective total joint replacement in persons with end-stage knee osteoarthritis. BMC Musculoskelet Disord 2010; 11: 86.
   
2. Kurtz S, Ong K, Lau E, Mowat F, Halpern M. Projections of primary and revision hip and knee arthroplasty in the United States from 2005 to 2030. J Bone Joint Surg Am 2007; 89: 780-5.
   
3. Parvizi J, Nunley RM, Berend KR, Lombardi AV Jr, Ruh EL, Clohisy JC, et al. High level of residual symptoms in young patients after total knee arthroplasty. Clin Orthop Relat Res 2014; 472: 133-7.
   
4. Beswick AD, Wylde V, Gooberman-Hill R, Blom A, Dieppe P. What proportion of patients report long-term pain after total hip or knee replacement for osteoarthritis? A systematic review of prospective studies in unselected patients. BMJ Open 2012; 2: e000435.
   
5. Kim MS, Kim JJ, Kang KH, Kim MJ, In Y. Diagnosis of central sensitization and its effects on postoperative outcomes following total knee arthroplasty: a systematic review and meta-analysis. Diagnostics (Basel) 2022; 12: 1248.
   
6. Zapparoli L, Sacheli LM, Seghezzi S, Preti M, Stucovitz E, Negrini F, et al. Motor imagery training speeds up gait recovery and decreases the risk of falls in patients submitted to total knee arthroplasty. Sci Rep 2020; 10: 8917.
   
7. Briones-Cantero M, Fernández-de-Las-Peñas C, Lluch-Girbés E, Osuna-Pérez MC, Navarro-Santana MJ, Plaza-Manzano G, et al. 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 Med 2020; 21: 3548-55.
   
8. Moukarzel M, Di Rienzo F, Lahoud JC, Hoyek F, Collet C, Guillot A, et al. The therapeutic role of motor imagery during the acute phase after total knee arthroplasty: a pilot study. Disabil Rehabil 2019; 41: 926-33.
   
9. Roy JS, Bouyer LJ, Langevin P, Mercier C. Beyond the joint: the role of central nervous system reorganizations in chronic musculoskeletal disorders. J Orthop Sports Phys Ther 2017; 47: 817-21.
   
10. Riquelme-Hernández C, Reyes-Barría JP, Vargas A, Gonzalez-Robaina Y, Zapata-Lamana R, Toloza-Ramirez D, et al. Effects of the practice of movement representation techniques in people undergoing knee and hip arthroplasty: a systematic review. Sports (Basel) 2022; 10: 198.
    
11. Moseley GL, Butler DS, Beames TB, Giles TJ. The graded motor imagery handbook. Noigroup Publications. 2012.
    
12. Moseley GL. A pain neuromatrix approach to patients with chronic pain. Man Ther 2003; 8: 130-40.
    
13. Strauss S, Barby S, Härtner J, Pfannmöller JP, Neumann N, Moseley GL, et al. Graded motor imagery modifies movement pain, cortical excitability and sensorimotor function in complex regional pain syndrome. Brain Commun 2021; 3: fcab216.
    
14. Moseley GL. Is successful rehabilitation of complex regional pain syndrome due to sustained attention to the affected limb? A randomised clinical trial. Pain 2005; 114: 54-61.
    
15. Dilek B, Ayhan C, Yagci G, Yakut Y. Effectiveness of the graded motor imagery to improve hand function in patients with distal radius fracture: a randomized controlled trial. J Hand Ther 2018; 31: 2-9.e1.
    
16. Moukarzel M, Guillot A, Di Rienzo F, Hoyek N. The therapeutic role of motor imagery during the chronic phase after total knee arthroplasty: a pilot randomized controlled trial. Eur J Phys Rehabil Med 2019; 55: 806-15.
    
17. Louw A, Schmidt SG, Louw C, Puentedura EJ. Moving without moving: immediate management following lumbar spine surgery using a graded motor imagery approach: a case report. Physiother Theory Pract 2015; 31: 509-17.
    
18. Dudhani S, Anwar KAM, Jain PK. Can brain cure pain and fear? Effect of graded motor imagery on post operative lumbar degenerative diseases - randomized control trial. Indian J Physiother Occup Ther 2020; 14: 219-26.
    
19. Birinci T, Kaya Mutlu E, Altun S. The efficacy of graded motor imagery in post-traumatic stiffness of elbow: a randomized controlled trial. J Shoulder Elbow Surg 2022; 31: 2147-56.
    
20. Lee HG, An J, Lee BH. The effect of progressive dynamic balance training on physical function, the ability to balance and quality of life among elderly women who underwent a total knee arthroplasty: a double-blind randomized control trial. Int J Environ Res Public Health 2021; 18: 2513.
    
21. Lagueux E, Charest J, Lefrançois-Caron E, Mauger ME, Mercier E, Savard K, et al. Modified graded motor imagery for complex regional pain syndrome type 1 of the upper extremity in the acute phase: a patient series. Int J Rehabil Res 2012; 35: 138-45.
    
22. Bade MJ, Stevens-Lapsley JE. Early high-intensity rehabilitation following total knee arthroplasty improves outcomes. J Orthop Sports Phys Ther 2011; 41: 932-41.
    
23. Carlsson AM. Assessment of chronic pain. I. Aspects of the reliability and validity of the visual analogue scale. Pain 1983; 16: 87-101.
    
24. Bellamy N, Buchanan WW, Goldsmith CH, Bellamy N, Goldsmith CH, Campbell J, et al. Validation study of WOMAC: a health status instrument for measuring clinically important patient relevant outcome following total hip or knee arthroplasty in osteoarthritis. J Orthop Rheumatol 1988; 1: 95-108.
25. Norkin CC, White DJ. Measurement of joint motion: a guide to goniometry. 5th ed. F.A. Davis Company. 2016.
    
26. Daloia LMT, Leonardi-Figueiredo MM, Martinez EZ, Mattiello-Sverzut AC. Isometric muscle strength in children and adolescents using Handheld dynamometry: reliability and normative data for the Brazilian population. Braz J Phys Ther 2018; 22: 474-83.
    
27. Podsiadlo D, Richardson S. The timed "Up & Go": a test of basic functional mobility for frail elderly persons. J Am Geriatr Soc 1991; 39: 142-8.
    
28. Bek J, Humphries S, Poliakoff E, Brady N. Mental rotation of hands and objects in ageing and Parkinson's disease: differentiating motor imagery and visuospatial ability. Exp Brain Res 2022; 240: 1991-2004.
    
29. Beauchet O, Annweiler C, Assal F, Bridenbaugh S, Herrmann FR, Kressig RW, et al. Imagined Timed Up & Go test: a new tool to assess higher-level gait and balance disorders in older adults? J Neurol Sci 2010; 294: 102-6.
    
30. Dilek B, Ayhan C, Yakut Y. Reliability and validity of the Turkish version of the movement imagery questionnaire-3: its cultural adaptation and psychometric properties. Neurol Sci Neurophysiol 2020; 37: 221-7.
    
31. Pelletier R, Higgins J, Bourbonnais D. Laterality recognition of images, motor performance, and aspects related to pain in participants with and without wrist/hand disorders: an observational cross-sectional study. Musculoskelet Sci Pract 2018; 35: 18-24.
    
32. Sullivan MJ, Bishop SR, Pivik J. The pain catastrophizing scale: development and validation. Psychol Assess 1995; 7: 524-32.
    
33. Süren M, Okan I, Gökbakan AM, Kaya Z, Erkorkmaz U, Arici S, et al. Factors associated with the pain catastrophizing scale and validation in a sample of the Turkish population. Turk J Med Sci 2014; 44: 104-8.
    
34. Mayer TG, Neblett R, Cohen H, Howard KJ, Choi YH, Williams MJ, et al. The development and psychometric validation of the central sensitization inventory. Pain Pract 2012; 12: 276-85.
    
35. Düzce Keleş E, Birtane M, Ekuklu G, Kılınçer C, Çalıyurt O, Taştekin N, et al. Validity and reliability of the Turkish version of the central sensitization inventory. Arch Rheumatol 2021; 36: 518-26.
    
36. Tunca Yılmaz Ö, Yakut Y, Uygur F, Uluğ N. [Turkish version of the Tampa Scale for Kinesiophobia and its test-retest reliability]. Fizyoter Rehabil 2011; 22: 44-9. Turkish.
37. Fritz CO, Morris PE, Richler JJ. Effect size estimates: current use, calculations, and interpretation. J Exp Psychol Gen 2012; 141: 2-18. Erratum in: J Exp Psychol Gen 2012; 141: 30.
    
38. Toms AD, Mandalia V, Haigh R, Hopwood B. The management of patients with painful total knee replacement. J Bone Joint Surg Br 2009; 91: 143-50.
    
39. Lewis GN, Parker RS, Sharma S, Rice DA, McNair PJ. Structural brain alterations before and after total knee arthroplasty: a longitudinal assessment. Pain Med 2018; 19: 2166-76.
    
40. Paravlic AH, Maffulli N, Kovač S, Pisot R. Home-based motor imagery intervention improves functional performance following total knee arthroplasty in the short term: a randomized controlled trial. J Orthop Surg Res 2020; 15: 451.
    
41. Gurudut P, Godse AN. Effectiveness of graded motor imagery in subjects with frozen shoulder: a pilot randomized controlled trial. Korean J Pain 2022; 35: 152-9.
    
42. Yap BWD, Lim ECW. The effects of motor imagery on pain and range of motion in musculoskeletal disorders: a systematic review using meta-analysis. Clin J Pain 2019; 35: 87-99.
    
43. Skoffer B, Dalgas U, Mechlenburg I, Søballe K, Maribo T. Functional performance is associated with both knee extensor and flexor muscle strength in patients scheduled for total knee arthroplasty: a cross-sectional study. J Rehabil Med 2015; 47: 454-9.
    
44. Spruijt S, van der Kamp J, Steenbergen B. Current insights in the development of children's motor imagery ability. Front Psychol 2015; 6: 787.
    
45. Paravlic AH, Pisot R, Marusic U. Specific and general adaptations following motor imagery practice focused on muscle strength in total knee arthroplasty rehabilitation: a randomized controlled trial. PLoS One 2019; 14: e0221089.
    
46. Montuori S, Curcio G, Sorrentino P, Belloni L, Sorrentino G, Foti F, et al. Functional role of internal and external visual imagery: preliminary evidences from pilates. Neural Plast 2018; 2018: 7235872.
    
47. Breckenridge JD, Ginn KA, Wallwork SB, McAuley JH. Do people with chronic musculoskeletal pain have impaired motor imagery? A meta-analytical systematic review of the left/right judgment task. J Pain 2019; 20: 119-32.
    
48. Anderson B, Meyster V. Treatment of a patient with central pain sensitization using graded motor imagery principles: a case report. J Chiropr Med 2018; 17: 264-7.
    
49. Louw A, Puentedura EJ, Reese D, Parker P, Miller T, Mintken PE. Immediate effects of mirror therapy in patients with shoulder pain and decreased range of motion. Arch Phys Med Rehabil 2017; 98: 1941-7.
    
50. Araya-Quintanilla F, Gutiérrez-Espinoza H, Jesús Muñoz-Yanez M, Rubio-Oyarzún D, Cavero-Redondo I, Martínez-Vizcaino V, et al. The short-term effect of graded motor imagery on the affective components of pain in subjects with chronic shoulder pain syndrome: open-label single-arm prospective study. Pain Med 2020; 21: 2496-501.