Many previous studies have compared the muscle activities between individuals with and without UT pain.[5, 20, 29–31] However, this is the first reported study to compare muscle activities, the EMG activity ratio, and the relative contribution of primary and synergic muscles during isometric shoulder abduction under different loads (low [25%], middle [50%], and high [75%]) between groups with and without UT pain. The present study investigated the decreased EMG activity of the primary muscle (MD) and altered recruitment patterns of scapular upward rotators in the UT pain group.
This study investigated altered recruitment patterns in the UT pain group. The EMG activity ratio has been used to assess the relative synergistic motor patterns. In the present study, the EMG activity ratio of SA/UT was significantly higher in the UT pain group than in the control group at 25% low-loading condition (UT pain: 1.13, Control: 0.83, p < 0.025). In addition, the relative contribution of SA activity in the low-loading condition was significantly greater in the UT pain group than in the control group (UT pain: 25%, control: 19.5%, p < 0.025); however, there was no significant difference in the relative contribution of UT activity in this study (UT pain: 25.3%, Control: 25.2%, p > 0.05). UT pain with MTrPs seems to lead to a compensatory movement by increasing the relative contribution of SA activity towards force couple motion. Sahrmann (2002) mentioned altered recruitment patterns as motor element impairments of the nervous system. Changes in recruitment patterns can lead to the action of a synergic muscle to become more dominant than the action of the other synergistic muscles. Clinically it can include consistent recruitment of one muscle of force-couple synergists. The previous study has demonstrated that after the SA was fatigued by electrical stimulation, the muscle activities of other synergies such as the UT and infraspinatus were increased to compensate for SA dysfunction. As seen in previous studies, the findings of the present study postulate that MTrPs can cause functional dysfunction of the UT and can result in altered muscle recruitment patterns of scapular upward rotators. Andersen et al. (2008c) showed that UT activity using surface EMG was lower in the UT pain group than in the control group. When we observed motor control strategy in various loading conditions, the UT pain group showed that the EMG activity ratio of SA/UT was greater under the low-loading condition and decreased with increasing load. In contrast, the control group had a relatively consistent ratio of SA/UT regardless of the loading conditions (Fig. 4). The increased relative activity of the SA in the low loading condition represents a suboptimal motor behavior in motor system adaptation theory, in which activity can be redistributed within and between muscles to protect the tissues from pain involving changes in the motor system.
We also observed decreased primary muscle activity in the UT pain group. The MD activity (%MVIC) was significantly decreased in the low loading condition (25%) in the UT pain group compared to the control group (UT pain: 23.1, control: 33.2, p < 0.05). This result supports the findings of a previous study that investigated the decreased shoulder abduction force in the UT pain group compared to the control group when scapular elevation and upward rotation were restricted. Altered recruitment patterns of synergistic muscles may cause a movement in the direction of action of the SA muscle. The main standard movement action of the SA is scapular protraction and upward rotation at scapulothoracic joint. Whereas, the standard actions of UT, are not only scapular elevation and upward rotation but also scapular retraction. However when the action of the SA becomes more dominant than that of UT, it can cause the scapular protraction to move toward the glenohumeral joint. This would cause not only decreased dynamic glenohumeral joint stability during shoulder abduction, but also an altered optimal length -tension relationship of the MD muscle. A previous study reported that application of the scapular reposition test, which imparted a force to posteriorly tilt, externally rotate, and retract the scapula improved the length tension relationship of the scapular musculature. As a result, there was reduced pain and increased shoulder elevation strength. An altered scapular position during scapulohumeral movement can change recruitment patterns in scapular stabilizing muscles, which may injure the shoulder joint. 
It is unclear whether UT pain with MTrPs is a primary phenomenon that may influence people to decrease MD activity, or secondary phenomenon as a result of MD dysfunction. Decreased MD activity can further exacerbate altered recruitment patterns. Lin et al. (2006) reported that scapular upward rotation during arm elevation was greater in individuals with anterior shoulder tightness than in the control group. Therefore, decreased MD activity in the UT pain group might lead to increased recruitment of the motor unit of the scapular upward rotators through a compensatory movement to induce the same amount of shoulder abduction performance.[1, 36]
Most rehabilitation exercises for shoulder impingement syndrome have focused on increasing the EMG activity ratio SA/UT to prevent abnormal patterns of decreased SA and excessive UT activity [16, 37, 38]. However, for UT pain with MTrPs, there seems to be a different motor control strategy pattern compared to that in shoulder impingement syndrome. Our findings showed that UT pain with MTrPs had abnormal patterns with decreased MD and increased relative contribution of SA activity. Andersen et al. (2008) found that the lower activity of the painful UT might be associated with decreased shoulder abduction torque. It also highlighted that shoulder abduction and elevation training increased MD strength capacity, UT and MD activation as well as decreased pain in chronic UT muscle. [10, 39] The present study supports the importance of the contractile capacity of the UT and MD under low loads (25% load) during 60° of isometric shoulder abductions.
In this study, all significant differences between groups in EMG activity, EMG activity ratio, and relative contribution were observed only under low loading conditions (25%). Aasa et al. (2015) highlighted that the low-loading condition is useful for managing motor control and that low-load exercise focuses on muscle recruitment patterns and recovery of optimal movement patterns. Therefore, muscle activity tests in low-load conditions might help to determine motor control deficits such as altered muscle recruitment. Reed et al. (2016) reported that the activity of all muscles increased with load in healthy subjects.  The present study demonstrated that the UT pain group also significantly increased motor unit firing in all muscles during a graded contraction (25%, 50%, and 75%); this was also seen in the healthy control group.
As shown in Table 3, the relative contribution of the UT increased with an increase in load, but that of the LT decreased with an increase in load. The EMG ratio of LT/UT was also significantly lower at 75% load than at 50% load (p < 0.016) (Fig. 4). This means that UT can be overloaded above a 50% load. Previous studies demonstrated that LT activity was significantly increased above 60º in the impingement group.  Neuman (2013) noted that LT activity was low at low angles but rapidly increased at high angles. Although LT activity had a tendency for high activity, by increasing load at 60° of shoulder abduction angle in both groups, no significant difference was observed between the UT pain group and the control group in this study. This finding suggests that increased LT activity may have been used as a stabilizer of the scapula rather than a scapular upward rotator as the load increases. However, angle seems to be an important factor in comparing LT activity. Further studies should compare the activity patterns of the LT at angles above 60º to investigate the differences in LT activity between the UT pain and control groups.
This study has several limitations. First, the activity of other scapular stabilizers, such as the rotator cuff muscles, levator scapulae, and rhomboids, were not investigated. The activity of the levator scapulae should be investigated to determine the compensatory movement of scapular elevation with UT. Second, shoulder abduction was performed only at 60° with isometric measurements to avoid the end range of the GH joint. The muscle acitivity patterns can be different during shoulder abductions of over 90°, in the UT pain group. Further studies are needed to determine the effects of low-load exercise to normalize altered recruitment patterns on decreasing UT pain in individuals with UT pain and MTrPs.