Effect of the Unilateral Breast Cancer Surgery on the Shoulder Movement: Electromyographic and Motion Analysis

After breast cancer surgery, women might develop musculoskeletal impairments that affect movements of the upper limbs and reduce the quality of life. The objective of this research was to analyze the effect of the unilateral breast cancer surgery on the kinematics of the shoulder and the electrical activity of the upper trapezius, middle deltoid and pectoralis major muscles. Eight right-handed female participants, mean age 46.5 ± 5.45 years and mean body mass 71.21 ± 13.33 kg, with unilateral breast cancer surgery, without breast reconstruction and without lymphedema symptoms were included in the research. Flexion-extension and abduction-adduction movements of the shoulder were evaluated with infrared cameras and the electrical activity was measured using surface electromyography. The statistical analysis of the direction angles showed a significant reduction of the flexion-extension and abduction-adduction movements in the affected side in most of the participants (p<0.05). The muscle electrical activity did not present a significant difference between the two sides for the flexion-extension and abduction-adduction movements (p>0.05). The results suggest that the surgical procedure could compromise the range of motion of the affected side. Furthermore, this research contributes to clarify the effect of the surgical procedure in the range of motion of the upper limbs.


INTRODUCTION
Breast cancer is the most common type of cancer in women in the world, with an incident of 1.7 million of new cases in 2016 [1] . For women, it was the most common cancer in 131 countries and cause of cancer deaths in 112 countries [1] [2] [3] [4] [5] [6] . In 2018 the International Agency for Research on Cancer estimated 2.08 million of new breast cancer cases worldwide [7] .
After breast cancer surgery, women might develop musculoskeletal impairments that affect movements of the upper limbs and reduce the quality of life.
Complications following breast cancer surgery include infection, swelling, hematoma, seroma and psychological factors such as, anxiety or depression [3] [8] [9] [10] [11] . The complications are not always localized on the region of the surgery, as many of the treatments have larger regional systemic effects on body structures and functions [3] [8] [11] [12] [13] . Although women report upper limb symptoms between 6 months and 3 years after breast cancer, some of the comorbidities could remain for a longer-term. The etiology of morbidity seems to be multifactorial, with the most consistent risk factors being those associated with extension of cancer treatment [14] .
The mastectomy produces dysfunctions on the upper limbs in a mid or long term, causes lymphedema, persistent pain [9] [15] , weakness and restricts the movement of the shoulder [10] [14] [16] . Furthermore, it has been found that muscle activation can be affected after breast cancer surgery. The effect depends on the type of surgery and type of reconstruction of the breast [17] [18] [19] [20] [21] . It has been thought that the most frequent complications include long-term weakening of muscles within the shoulder and upper limb at the affected side [4] . It is believed that if one muscle is compromised, then other muscles might become more active to compensate for the lost movement. However, the muscle electrical activity can be influenced by age, dominance or pain on the limb [20] .
Movements of the upper limbs and head are recommended as rehabilitation after a breast cancer surgery.
Upper-body morbidity may be treatable with physical therapy. It has been found a 50-53% reduction in the risk of breast cancer deaths in women who are physically active after a breast cancer diagnosis [6] [14] .
Although rehabilitation represents an alternative to reduce the complication after breast cancer surgery, many of the patients do not receive the appropriate treatment to address the complications [22] [23] [24] .
The complications after mastectomy are well known, evaluated and treated from measurements such as goniometry, observation and strength tests with manual resistance [8] [10] . However, they do not provide specific scenarios for kinematic discontinuities of the shoulder movement. Moreover, there are few studies focused on the analysis of the biomechanics of the upper limbs on a three-dimensional space after breast cancer treatment [20] [25] [26] [27] . Biomechanical and muscular activation studies with equipment with higher accuracy are required to recognize these affectations and identify areas of opportunity to improve physiotherapeutic intervention plans. Therefore, the objective of this research is to analyze the movement of the shoulder and the electrical activity of the trapezius, deltoid and pectoralis major muscles in patients with unilateral breast cancer surgery.

Experimental protocol
Eight right-handed female participants, mean age 46.5 ± 5.45 years and mean body mass 71.21 ± 13.33 kg with unilateral breast cancer surgery (mastectomy), without breast reconstruction and without lymphedema symptoms after the surgery were included in the research, see Table 1. The sampled population was selected with non-probabilistic sampling. After the surgery, all the participants performed a rehabilitation process of the upper limbs. This rehabilitation was performed for 60 days and involved series of active movements of the head, flexion-extension and abduction-adduction of the arms. Participants which could not follow instruction, presented post-surgery issues or did not complete the rehabilitation were excluded. Simultaneously, personal data were recorded and saved on an encrypted computer for safekeeping.
Then, the participants performed three trials of bilateral flexion-extension and abduction-adduction movement of the arms. Each trial consisted of ten cycles performed over a period of 20 seconds paced using a metronome. Three minutes were allowed among trials to give rest to the patients and avoid fatigue.

Data processing
The data of the markers were exported to Matlab The raw EMG data was filtered with a 2-pole zero-lag Butterworth band pass filter with cut-off frequencies of 5 and 400Hz in order to retain as much as possible of the electrical activity data and reject high frequency noise. The cut-off frequencies were obtained from a frequency spectrum analysis of the raw EMG data [28] .
Then, the root mean square (RMS) value of every muscle burst for all the trials was calculated within a window of 100ms. The RMS window was calculated at the maximum peak of each burst (50ms forwards and 50ms backwards). Finally, the RMS values of the muscles were averaged and then, the healthy side was compared versus the affected side.

Statistical analysis
In order to determine the effect of the unilateral breast cancer surgery on the shoulder, a Student's t-test for dependent samples was performed. The comparison of the direction angles and the electrical activity of the muscles was done for both sides. The significance level was taken to be p < 0.05.

Flexion-Extension movement
The direction angles from Figure 2a show the movement of the arms during the flexion-extension motion (subject 8). As the movement is executed mainly in the sagittal plane, the direction angle of the X axis (alpha) does not change significantly. However, the movement of the arms with respect to the Y and Z axes (beta and gamma) changes. Figure 2a shows that the affected arm (subject 8-right arm) reduces the direction angles (beta and gamma) when the arm is reach-  Figure 2a.  Table 2 shows the mean and standard deviation values (alpha, beta and gamma) of the eight subjects during the flexion-extension motion of both arms. In general, it was found that the affected side presented less amplitude of motion, excluding subject 5. The statistical analysis showed a significant reduction of the movement in the affected side (p < 0.05).  The statistical analysis showed a significant reduction of the movement in almost all affected sides (p < 0.05).

Muscle electrical activity during the flexion-extension movement
For flexion-extension movement, the trapezius showed the highest electrical activity followed by the deltoid muscle. Although the electrical activity was higher in some cases, the statistical analysis did not show a significant difference, as shown in Table 3 (p > 0.05). The trapezius in subject one was the only muscle with a significant statistical difference between the affected and non-affected side (p = 0.047). However, there was not a specific trend of the data to justify that the affected side increases or decreases the electrical activity after the rehabilitation post-surgery.

Muscle electrical activity during the abduction-adduction movement
The RMS and the standard deviation values of the three muscles during the abduction-adduction movement are registered in Table 3. Although it was expected to find a significant difference among the muscles, the statistical analysis did not show a significant effect. The electrical activity of the trapezius muscle in subject one presented a significant statistical difference (p = 0.043). However, it could not be established a specific trend of the EMG behavior. [14] [27] [29] . Furthermore, the reduction of the ROM could restraint the performance of some activities and then, have an impact on the quality of life of the patients [8] [9] [10] [26] . It has been mentioned that the scar tissue formation and protecting posturing after the breast cancer surgery could lead to shortening of the anterior chest wall. Therefore, the reduction of the ROM of the upper limbs found in this study could be related to the shortening of the breast chest [13] [16] .
Some studies have found a significant reduction in the ROM of the upper limbs after breast cancer surgery using goniometry [8] [10] [17] . Our study used infrared strategies and try to cope with the morbidities [10] [11] .
The electrical activity of the muscles did not present a significant difference between the affected and non-affected side. Perhaps this could be produced due to the fact that the patients unconsciously try to elevate the upper limbs without too much effort.
Furthermore, from the anthropometric data (Table 1), most of the participants were overweight or obese. We believe that overweight or obese participants will have different muscle electrical activity than the normal weight participants as the fat tissue could attenuate the EMG signal. This could be hidden the real effect of the muscle electrical activity when the affected and non-affected sides are compared. On the other hand, it has been suggested that humeral elevation of the arm, upward movement on the affected side, and left shoulder movement will increase the electrical activity of the muscles regardless of which side is affected [17] . Some studies have found significant difference in the EMG activity and muscle strength reduction after breast cancer surgery [19] [21] [30] . Although our study considers patients without pain, all of them were treated with coadjutant therapy (chemotherapy), which could have an effect on the electrical activity [18] [21] . Furthermore, it has been found that the EMG activity depends on parameters such as age, dominant hand, affected side, type of surgery, abnormal activation patterns and breast cancer treatment [17] .
It has been recognized that exercise training part of the rehabilitation program will reduce the risk of breast cancer deaths in women who are physically active after breast cancer diagnosis [6] . Therefore, it is recommended to continue with the practice of physical therapy for improving the ROM of the affected side [6] .  [31] .
The current study presents some limitations.