The upper aero-digestive tract is subject to a wide variety of head and neck cancers (HNCs), which are most commonly caused by alcoholism, tobacco use and human papillomavirus (HPV) infection. Such neoplasms typically originate in the mucosal lining [1], and can develop into squamous cell carcinoma (SCC) in the lip, mouth, salivary glands, larynx, thyroid and nasopharynx [2]. SCC is considered the seventh most prevalent cancer globally and can develop in various subsites [3]. The WHO classifies the various forms of head and neck malignancies according to their anatomical location [4].
One surgical technique for treating cervical lymph node metastases is neck dissection (ND), which can be used to remove cancerous growths that have spread to the cervical lymph nodes [5]. ND is regarded as the foremost procedure to treat cervical lymph node cancers for diagnosis (staging) and treatment [6]. However, ND can result in spinal accessory nerve damage, resulting in chronic neck pain, shoulder dysfunction, trapezius atrophy, and scapular dyskinesia [7].
A common complication following NDS is accessory nerve shoulder dysfunction (ANSD), caused by weakness of the trapezius muscle due to accessory nerve (AN) denervation or resection. Its common symptoms include pain, heaviness, shoulder protrusion, depression, and limited range of motion (ROM), particularly abduction and winging of the scapula [8]. ANSD results in decreased elevation, rotation, and abduction of the shoulders, as well as their dropping and decreased range of motion (ROM) [9]. Even when attempts to preserve AN are made, up to 40% of patients report having problems with their shoulders after surgery, and substantial palsy can limit shoulder abduction to between 90° and 120° [10].
In such cases, the peripheral joints can be treated by applying osteo-kinematic movements and sustained gliding forces to the joint in order to realign the bony positional defects; a procedure known as Mulligan mobilization with movement (MWM) [11]. In MWM, shoulder dysfunction is treated by applying passive glenohumeral joint (GHJ) joint gliding movements, followed by extra pressure when at the limit of ROM, and then an active or passive range of motion that is pain-free [12]. Such treatment allows anti-inflammatory mediators to inhibit pain nociceptors and correct minor shoulder joint position faults, thus improving joint mobility and mitigating pain [13]. MWM decreases painful restriction of shoulder movement, as forward translation of the humeral head can cause a positional fault; however, this forward translation is corrected by the posterolateral glide, increasing the shoulder ROM [14].
The scapulothoracic joint supports the glenohumeral joint while it moves, allowing the shoulder to move in synergy with the trunk by providing the scapula has six degrees of freedom and allowing it to move in three dimensions (upward and downward rotation, internal and external rotation, and anterior/posterior tilt). In the absence of proper scapular alignment and arthrokinematics, impaired periscapular muscle function can ultimately lead to the development of overuse shoulder pathology or secondary injury, reduced shoulder ROM and lower shoulder pain and disability index (SPADI) scores; conversely, higher SPADI and greater ROM indicate greater scapulothoracic stability and alignment [15].
Previous systematic reviews assessing the efficacy of MWM have lacked population-specific studies of the shoulder and those describing precise work on the shoulder. Also, little is known of the advantages of MWM over other non-surgical management techniques, such as electrotherapy, placebo, or sham therapy [16]. The aim of the study is therefore to assess the therapeutic impact of the MWM on enhancing shoulder ROM and upper extremity function in patients recovering from shoulder dysfunction following neck dissection surgery.
A double-blind randomized controlled trial with controls was conducted to assess the effectiveness of MWM in alleviating shoulder dysfunction following neck dissection surgery. The participants were randomly assigned to groups using a computer-generated block randomization program. The study was conducted between July 2023 and April 2024 at Damanhour Oncology Center, Damanhour, El-Behira, Egypt.
The study received approval from the Research Ethics Committee at the Faculty of Physical Therapy, Cairo University, Egypt (Approval No.: P.T.REC/012/004500). It was performed in accordance with the Declaration of Helsinki regarding research on human subjects. All participants were informed about the goals, design, and advantages of the research, and that they were free to leave at any moment and without explanation. Before the study began, each participant had to provide their signed, written consent to take part. The investigation was documented on ClinicalTrials.gov. (Registration No.: NCT05915572).
Eighty-six patients with NDs were referred to the study by an Oncology surgeon at Damanhur Oncology Center. Of these, ten were not eligible to participate: four who refused to take part in the study, and six did not meet the inclusion criteria. Therefore, seventy-six patients experiencing shoulder dysfunction following neck dissection surgeries were included in the study (Figure 1).
CONSORT flow chart of the study
The 76 participants were randomly allocated into two similar-sized groups using SPSS software for Windows, version 25 (SPSS Inc, Chicago, USA). To blind the allocation, the randomization codes were safely kept in sealed, transparent envelopes with sequential numbers. Randomization was carried out by an evaluator who was not involved in the therapeutic interventions or the evaluation.
The inclusion criteria were as follows: (1) age between 30 and 50 years; (2) shoulder pain and disability following modified radical neck dissection (MRND) procedures; (3) pain at moderate to severe level (VAS > 4); (4) ability to understand instructions clearly; (5) provide informed consent. The exclusion criteria comprised (1) rotator cuff tears; (2) injuries to the shoulder ligaments; (3) cancers in the shoulder area; (4) rheumatoid arthritis; (5) shoulder fracture; (6) shoulder dislocation; (7) GH joint musculoskeletal injuries; (8) neurological disorders; (9) fracture or spondylolisthesis of the cervical spine; (10) tendon calcification; (11) injuries to peripheral nerves; (12) seizures; (13) cervical disk; (14) cervical radiculopathy; and (15) thoracic outlet syndrome.
The primary outcome measure was shoulder active ROM (flexion, abduction, and external rotation), while the secondary outcome measure was upper limb function. Shoulder active ROM was evaluated by a digital goniometer, and upper limb function was evaluated by the shoulder pain and disability index (SPADI). Both were assessed prior to and following the treatment.
It is an effective tool for accurately measuring joint ranges, velocity, and acceleration parameters with precision and affordability comparable to conventional analog instruments. As it is specifically designed for measuring body movement angles with a resolution of 1°, it offers superior results than traditional goniometers that typically use 5° increment scales [17]. To measure the selected active shoulder movements, the patient was instructed to recline in a comfortable supine position, with the elbow extended, thumb pointing upward, and arm next to the body [18].
The SPADI demonstrated high reliability, indicated by a Cronbach's alpha value of 0.920, measuring internal consistency, Test-retest reliability was high with an intra-class coefficient (lCC) of 0.935, with a 95% confidence interval of 0.887–0.963. It also had acceptable criterion validity, as indicated by a moderately-negative Pearson's correlation coefficient (r = −0.457 to −0.683). However, the visual analogue scale (VAS) and SPADI have a strong positive correlation for both pain (r = 0.871, 0.910, 0.891) and disability (r = 0.877, 0.915, and 0.813), indicating strong convergent validity, when a significant difference (p < 0.05) was found between the groups, the Mann-Whitney U-test was employed to assess discriminant validity [19].
The SPADI is a self-reported questionnaire consisting of 13 items divided into two parts: the pain and disability subscales. The pain subscale consists of five questions concerning the worst possible pain: lying on the affected side, touching the back of the neck, reaching up on a high shelf, and pushing with the affected arm. The disability subscale consists of eight questions about challenges with hair washing, back washing, putting on a jumper or undershirt, putting on a shirt with buttons down the front, putting on pants, placing an object on a high shelf, carrying a ten-pound (4.5 kg) object, and taking something out of the back pocket. Each item on the subscale measuring pain and disability is rated from 0 to 10. The scores for the pain and disability subscales are totaled and converted to percentages, with the final score being the mean values. A score of 0% indicates the best result and 100% the worst [20].
In Group A (38 patients), the participants received Mulligan mobilization three times per week in addition to a six-week conventional physiotherapy program comprising myofascial release, stretching, strengthening, and ROM exercises. In Group B (38 patients) the participants only received the six-week conventional physiotherapy program comprising myofascial release, stretching, strengthening, and ROM exercises. The therapy involved 18 sessions scheduled over a period of six weeks, with three sessions each week.
Every patient in the study received complete details about the treatment program and the advantages of the MWM to motivate them during treatment. In Group A, the affected shoulder was treated using Mulligan mobilization with movement technique (MWM) pioneered by Brian Mulligan: a form of manual therapy for peripheral joints that involves a combination of an active movement with simultaneous passive accessory mobilization that aims to achieve painless range of motion by restoring the reduced accessory glide [21]. The treatment was performed by a physiotherapist with experience in the MWM. During the sessions, passive accessory glide was applied, while the patient actively moved their arm within a pain-free range in flexion, abduction, and external rotation. Each exercise was performed as three sets of 10 repetitions, with a total duration of 15–30 mins. Additionally, the participants received a conventional physiotherapy program comprising myofascial release (10 mins), stretching (10 mins), strengthening (15 mins), and ROM exercises (10 mins); total duration of session: 60–75 mins; frequency: three sessions per week; duration of treatment: six weeks; total number of sessions: 18 sessions [22].
1-MWM to increase shoulder flexion: the patient begins in a relaxed sitting position, with the therapist in a stride standing position, standing laterally from the affected joint. The therapist then uses one hand to stabilize the scapula and clavicle, and the other to support the distal end of the humerus. The patient is instructed to perform passive overpressure at the end of the new ROM with a slow active shoulder movement (flexion) using a belt fastened around the therapist's waist and placed close to the line of the shoulder joint; this is performed while maintaining a posterolateral glide to distract the humerus laterally. In order to maintain the glide along the treatment plane, the therapist moves with the patient.
2-MWM to increase shoulder abduction: the patient begins in a relaxed sitting position, and the therapist is stride standing. A belt is fastened around the humeral head, ensuring that the posterior-lateral and inferior movements are maintained. The therapist maintains the glide by holding the belt in place with one hand. The therapist also uses the other hand to apply counter pressure toward the scapula. The patient is instructed to actively and gradually move the affected shoulder to the limit of its pain-free range (abduction). Once the desired movement is completed and the therapist returns to the starting position, the glide is released.
3-MWM to increase shoulder external rotation: the patient begins in a supine position, with the therapist stride standing, standing laterally from the affected joint. The patient's elbow and shoulder are flexed to 90 degrees, and the therapist's belt is secured around the patient's waist near the line of the affected shoulder joint. The therapist holds the distal end of the humerus using both hands. While applying lateral traction to the belt to distract and stabilize the joint, the therapist instructs the patient to actively perform the correct movements (internal and external rotation). Additionally, the therapist gently applies passive pressure at the end of the newly-achieved ROM with the other hand [23].
The number of participants needed for the study was calculated using G*Power 3.1.9.4 software (Heinrich-Heine-University, Düsseldorf, Germany). For a two-tailed test comparing means between two independent groups, the computation of sample size depended on a type I error (α = 0.05), power (1-β = 0.85), and an effect size (d = 0.7) for the main variable outcome. The suitable size determined for the study was 76 patients, with 38 patients allocated to each group.
Age, BMI, and time since surgery were compared between groups using an unpaired t-test, and sex, affected side, and type of treatment with a chi-squared test. The normal distribution of the data was tested using the Shapiro-Wilk test. Homogeneity among groups was confirmed using Levene's test. The influence of intervention on shoulder flexion, abduction, external rotation ROM, and SPADI was confirmed using mixed MANOVA. The significance level for all statistical tests was established at p < 0.05. Statistical analyses were conducted using SPSS Statistics, version 25 for Windows (IBM SPSS, Chicago, IL, USA).
The attributes of the participants in groups A and B are given in Table 1. No significant differences in the distribution of sex, age, BMI, time since surgery, affected side (side of ND), or type of intervention were found between the groups.
Characteristics of participants in Group A and Group B, and a comparison between groups
| Group A | Group B | |||
|---|---|---|---|---|
| Mean ± SD | Mean ± SD | t-value | p-value | |
| Age [years] | 43.21 ± 6.01 | 42.39 ± 5.41 | 0.62 | 0.54 |
| BMI [kg/m2] | 28.29 ± 2.05 | 27.74 ± 2.24 | 1.12 | 0.27 |
| Time since surgery [months] | 1.75 ± 0.72 | 1.97 ± 1.03 | −1.08 | 0.28 |
| Sex, N (%) | ||||
| Women | 26 (68%) | 24 (63%) | (χ2 = 0.23) | 0.63 |
| Men | 12 (32%) | 14 (37%) | ||
| Side of ND, N (%) | ||||
| Dominant side | 21 (55%) | 20 (53%) | (χ2 = 0.05) | 0.82 |
| Non dominant side | 17 (45%) | 18 (47%) | ||
| Type of treatment, N (%) | ||||
| Chemotherapy & Surgery | 14 (36.8%) | 11 (29%) | (χ2 = 0.97) | 0.61 |
| Radiotherapy & Surgery & Chemotherapy 12 (31.6%) | 16 (42%) | |||
| Surgery 12 (31.6%) | 11 (29%) |
p-value- probability value, SD- standard deviations, χ2- chi squared value, BMI- body mass index, ND- neck dissection, N- number
Mixed MANOVA showed a significant interaction between intervention and time (F = 181.47, p = 0.001, ηp 2 = 0.91). A significant main effect was noted for treatment (F = 106.24, p = 0.001, ηp 2 = 0.86) and for main effect over time of intervention (F = 806.44, p = 0.001, ηp 2 = 0.98). The interaction effect indicates that the applied intervention (MWM plus conventional physical therapy program) yielded a significant improvement in shoulder ROM (flexion, abduction and external rotation) and SPADI scores between pre and post-intervention (six weeks) (p < 0.001).
Both groups A and B demonstrated significant increases in shoulder flexion, abduction and external rotation ROM over the course of the intervention (p < 0.001). These respective increases were 19.21, 28.89, and 29.68% in Group A, and 7.86, 8.07, and 8.89% in Group B (Table 2).
Mean shoulder flexion, abduction, and external rotation ROM before and after intervention in groups A and B
| Shoulder ROM (degrees) | Group A | Group B | |||
|---|---|---|---|---|---|
| Mean ± SD | Mean ± SD | MD | p-value | Effect size | |
| Flexion | |||||
| Pre-treatment | 101.95 ± 14.23 | 100.13 ± 16.21 | 1.82 | 0.61 | |
| Post-treatment | 121.53 ± 12.34 | 108 ± 15.87 | 13.53 | 0.001 | 0.95 |
| MD | −19.58 | −7.87 | |||
| % of change | 19.21 | 7.86 | |||
| p = 0.001 | p = 0.001 | ||||
| Abduction | |||||
| Pre-treatment | 90.59 ± 11.84 | 88.76 ± 14.01 | 1.83 | 0.54 | |
| Post-treatment | 116.76 ± 12.81 | 95.92 ± 14.48 | 20.84 | 0.001 | 1.52 |
| MD | −26.17 | −7.16 | |||
| % of change | 28.89 | 8.07 | |||
| p = 0.001 | p = 0.001 | ||||
| External rotation | |||||
| Pre-treatment | 63.74 ± 8.51 | 64.21 ± 8.63 | −0.47 | 0.81 | |
| Post-treatment | 82.66 ± 5.37 | 69.92 ± 8.48 | 12.74 | 0.001 | 1.79 |
| MD | −18.92 | −5.71 | |||
| % of change | 29.68 | 8.89 | |||
| p = 0.001 | p = 0.001 |
MD- mean difference, p-value- probability value, SD- standard deviation, ROM- range of motion
In addition, significant decreases in SPADI were noted following the intervention (p < 0.001). The percentage decrease was 58.65 in Group A and 22.30% in Group B (Table 3).
Mean SPADI before and after intervention in groups A as well as B
| SPADI (%) | Group A | Group B | |||
|---|---|---|---|---|---|
| Mean ± SD | Mean ± SD | MD | p-value | Effect size | |
| Pre-treatment | 91.08 ± 4.37 | 91.34 ± 3.89 | −0.26 | 0.78 | |
| Post-treatment | 37.66 ± 5.48 | 70.97 ± 5.84 | −33.31 | 0.001 | 5.88 |
| MD | 53.42 | 20.37 | |||
| % of change | 58.65 | 22.30 | |||
| p = 0.001 | p = 0.001 |
MD- mean difference, p-value- probability value, SD- standard deviation, SPADI- shoulder pain and disability index
No appreciable differences were noted between the groups prior to treatment (p > 0.05). Following therapy, Group A had significantly less SPADI (effect size = 5.88) than group B (p < 0.001), as well as significantly more shoulder flexion (effect size = 0.95), abduction (effect size = 1.52), as well as external rotation (effect size = 1.79) ROM. (Table 2 and 3).
This study was designed to assess the therapeutic influence of the mulligan mobilization technique on improving active shoulder ROM and upper extremity function following neck dissection surgery. Both groups demonstrated a significant increase in active range of motion (p < 0.001), as well as a substantial (p < 0.001) decline in SPADI score, over the course of treatment. These indicate improvements in the state of the scapulothoracic and glenohumeral joint.
Our findings agree with those of Longo et al. [24], who found that the scapula plays a crucial role in stabilizing and preserving the structural integrity of the shoulder by limiting excessive translation during three-dimensional motions and supporting shoulder kinematics. They are also in accordance with Madhumita et al. [25], who investigated the impact of MWM and shoulder strengthening exercises on reducing pain and improving shoulder function among individuals with shoulder peri-arthritis: both groups A (MWM) and B (exercises for shoulder power) demonstrated significant differences between pre- and post-treatment (p= 0.0001), and that MWM significantly affects PPT recovery and functional activity in individuals with periarthritis of the shoulder joint. The study concluded that MWM improved functional activity and lowered the pain threshold to a significant extent.
Our findings also agree with those of a meta-analysis of randomized controlled studies by Shahzad et al. [26], who examined the influence of mobilization techniques on shoulder disorders. The primary outcomes of interest were pain levels and shoulder mobility improvements. A significant reduction in pain was noted during treatment (p < 0.05), as well as significant increases in arm ROM: toward and away from the body, as well as internal and external rotation (p ≤ 0.05), the majority of the literature indicates that participants undergoing mobilization therapy showed significant enhancements in shoulder function. Such techniques, including scapular mobilization and mulligan mobilization, are successful in enhancing shoulder disability and quality of life.
Our findings also confirm those of Elabd et al. [27], who studied the effectiveness of MET and MWM for improving upper extremity function and posture. The results found that MWM realigns the joints to restore proper joint arthrokinematics, ensuring they move correctly along their natural paths. The mechanical advantages might include improving the extensibility of the shoulder capsule, stretching tightened soft tissue, restoring the scapulohumeral rhythm, and improving the sliding of fibers in areas of the capsule that are stressed during specific movements. As such, so it can be concluded that MWM is significantly beneficial for shoulder ROM, posture, and impairment.
In addition, a study of the impact of MWM on discomfort, painful motion, and quality of life in patients experiencing shoulder pain and movement limitations found that MWM improved external rotation and shoulder abduction, MWM also improved various shoulder ROM in comparison to the Maitland technique, indicating that MWM could decrease shoulder pain and discomfort, and improve function and shoulder range of motion [28].
Similarly, a previous study compared the effects of Maitland and Mulligan mobilization on discomfort level, ROM and impairments among patients diagnosed with shoulder impairments post-NDS, MWM to be a joint mobilization technique that includes persistent gliding and active motion while the joint is bearing weight. It is based on the principle of joint mobilization with simultaneous patient-generated movement. The direction of force application in the MWM is selected based on the forces created in different joints, with most suitable and efficient force applied to increase shoulder range of motion. The study concluded that the Mulligan mobilization technique has significant effects on discomfort, ROM, and shoulder dysfunction [29].
Additionally, our findings confirm those of Nawaz et al. [30], who compared the influence of MWM and muscle energy techniques (MET) on shoulder impairments. Their results showed both interventions to yield significant variance in the NPRS, SPADI, and The Disabilities of the Arm, Shoulder and Hand questionnaire (DASH) scores, thus improving motion and quality of life in individuals with shoulder dysfunction. However, the MWM was found to be more effective than MET on reducing pain and increasing movement in patients with shoulder dysfunction as it showed more substantial results in decreasing impairment and limitations, and increasing range of motion.
Similarly, a systematic review by Pananwala et al. [31] found the Mulligan mobilization technique to yield significant improvements in terms of pain, discomfort, and movement disability in shoulder joint pathologies. The study included six research articles, including experimental prospective studies, systematic reviews, and randomized control trials. All of the selected articles found the most substantial improvement to be in flexion ROM.
While the spinal accessory nerve (SAN) is purposefully removed during radical neck dissection, it is preserved in modified-radical and selective neck dissections. Therefore, while SAN preservation is possible in some types of NDS, these surgical procedures often entail substantial nerve complications. SAN injury results in the development of shoulder syndrome, characterized by neuropathic pain, trapezius muscle paralysis, shoulder immobility, scapular winging, and shoulder drop [32].
MWM can improve shoulder ROM and upper limb function through its combination of simultaneous passive accessory mobilization with an active movement that reduces pain and increases range of motion by restoring the reduced accessory glide [33]. It has also been proposed that the MWM decreases pain by exerting neurophysiologic effects via nociceptor inhibition and peripheral mechano-receptor stimulation [34]. During Mulligan mobilization, mechanical force may be applied to break up adhesions, or increase fiber sliding. Moreover, mobilization methods in general aim to preserve or improve joint mobility by causing biological alterations in synovial fluid and improving exchange [35]. Since the MWM involves the therapist gently gliding over the joint while the patient actively moves the joint at the same time, it was assumed that MWM would be useful in the treatment of musculoskeletal disorders. Indeed, the foundation of the MWM, known as the joint tracking mechanism model. MWM procedures have also been found to be effective at decreasing painful discomfort and enhancing joint movements in the spine and peripheral joints [36].
No side effects were observed for the treatment. Nonetheless, the study has some limitations that should be taken into consideration. The data was acquired from. The cancer surgeons would also vary in their expertise and profession. Furthermore, some mistakes may have been made in the measurements taken during intervention. In addition, the psychological condition of the patients and their cooperation during treatment may also have influenced the results.
The combination of Mulligan mobilization technique and a conventional physiotherapy program were found to be effective in improving shoulder active ROM and upper limb function in patients with shoulder dysfunction following neck dissection surgery.