Compensatory activation of the deltoid and remaining rotator cuff muscles in patients with rotator cuff tears using real-time tissue elastography Authors: Kyosuke Hoshikawa, Ryuta Oishi, Takuma Yuri, Tomohiro Uno, Jun Nagai, Hugo Giambini, Nariyuki Mura Published: 2026 ABSTRACT AbstractIntroductionRotator cuff (RC) tears are a prevalent condition leading to dysfunction and shoulder pain. Spatial changes in the behavior of the deltoid and remaining RC muscles during arm elevation in patients with RC tears have not been investigated. The purpose of this study was twofold: (1) to investigate factors related to the changes in the deltoid muscle behavior during scapular plane abduction (scaption) in patients with RC tears, and (2) to examine the functional relationship between the deltoid and remaining RC muscles, particularly remaining infraspinatus (ISP) and teres minor (TMin) muscles, in these patients using ultrasound elastography.MethodsTwenty-four patients (mean age, 66 ± 6 years) with RC tears (mean antero-posterior diameter, 17.4 ± 8.3 mm) were evaluated. Antero-posterior and medio-lateral diameters of the posterosuperior RC and Subscapularis (SSC) tears were measured intra-operatively. Muscle elasticity was evaluated at passive and active states, as a proxy for muscle activation, at 30°, 60°, and 90° during scaption for the deltoid, ISP, and TMin muscles using real-time tissue elastography. Based on the deltoid muscle activity ratio at 30° and 90° scaption, patients were divided into normal and abnormal activation groups.ResultsActivity of the deltoid muscle in the abnormal activation group exhibited consistently high activity across various elevation angles, significantly higher compared to that of the normal activation group at 30° scaption (p = 0.037). Although there was no statistically significant between-group difference in TMin muscle activity at 30° scaption, TMin muscle activity in the normal activation group was highest at 30° and significantly higher than at 90° scaption (p = 0.032). Abnormal activation group had a higher incidence of SSC tears combined with posterosuperior RC tears (p = 0.009).ConclusionsSSC tears combined with posterosuperior RC tears may be associated with compensatory activation of the deltoid muscle in the early phase of arm elevation, while relatively greater TMin muscle activity in the initial range of elevation may reflect a compensatory contribution of the remaining posterior RC muscles.Keywords: Rotator cuff tears, Deltoid muscle, Remaining rotator cuff muscles, Compensatory activation, Real-time tissue elastography Introduction Rotator cuff (RC) tears are a prevalent condition leading to dysfunction and shoulder pain. Management of RC tears includes surgical or nonoperative treatments. However, the criteria for determining the appropriate treatment have traditionally been ambiguous. A Delphi questionnaire study reached consensus on several criteria for surgical referral, such as the severity of pain, functional limitation, identification of degeneration, and the necessity to attempt a course of nonoperative treatments before considering surgery [9]. Among these, functional limitation is a particularly significant factor in deciding whether to pursue surgical treatment. Normal shoulder elevation is achieved through the force couple balance between the deltoid and rotator cuff muscles in the coronal plane [4]. Biomechanical studies have shown that the loss of the abduction force in the torn RC muscles can be compensated by the remaining RC and deltoid muscles [7, 11, 20, 27]. Additionally, clinical studies have indicated increased activity in both the deltoid and remaining RC muscles during elevation in patients with RC tears [5, 12, 28]. de Witte et al. suggested that measuring the compensational activity of the deltoid muscle could serve as an estimate of the activity of the remaining RC muscles [5]. However, the precise functional relationship between the deltoid and RC muscles during shoulder elevation in patients with RC tears remains unclear. Ultrasound elastography is a novel technique for estimating muscle activity by measuring muscle stiffness, which is calculated through strain ratio (%) [32, 33] or shear wave modulus (kPa) [2, 3]. Some studies have investigated the functional relationship between the deltoid and supraspinatus (SSP) muscles using ultrasound elastography [17, 26, 30]. Ishikawa et al. observed the activity ratio between the deltoid and SSP muscles in patients following RC repair. They found that SSP muscle activity significantly increased from 6 weeks to 3 months post-surgery, reaching levels comparable to healthy controls, while deltoid muscle activity significantly decreased from 6 weeks to 6 months post-surgery [17]. Similarly, Sakaki et al. reported that the activity of the anterior-superficial region of the SSP muscle during shoulder abduction significantly increased at 12 months post-surgery compared to pre-surgery and 3 months post-surgery. In contrast, the activity of the deltoid and upper trapezius muscles significantly decreased at 12 months post-surgery compared to pre-surgery [26]. However, no studies have investigated the functional relationship between the deltoid and remaining RC muscles during elevation in patients with RC tears. In a pilot study, we observed two distinct patterns of deltoid muscle behavior within the same cohort of patients diagnosed with full thickness RC tears who were able to actively elevate their arms beyond 90° during shoulder abduction in the scapular plane (scaption). In one pattern, deltoid activity increased linearly with shoulder elevation angle, peaking at 90°, similar to that observed in healthy individuals [16]. In the other pattern, high deltoid activity was observed from the initial phase of elevation, with relatively constant activity throughout the range of elevation. We hypothesized that these differences in deltoid muscle behavior are influenced by the tear characteristics, particularly the involvement of the subscapularis (SSC) combined with posterosuperior tears, and by functional changes in the remaining RC muscles. Therefore, the purpose of this study was twofold: (1) to investigate factors related to the changes in the deltoid muscle behavior during scaption in patients with RC tears, and (2) to examine the functional relationship between the deltoid and remaining RC muscles, particularly remaining ISP and TMin muscles, using ultrasound elastography. Methods Patients Twenty-eight patients who underwent arthroscopic RC repair between September 2021 to March 2022 were recruited to this study. Inclusion criteria were as follows: (1) full-thickness RC tear identified on preoperative magnetic resonance imaging (MRI); (2) underwent arthroscopic RC repair. Exclusion criteria were as follows: (1) partial-thickness tear; (2) contracture of the affected shoulder, passive range of motion (ROM) during elevation < 100° under anesthesia [8]; (3) active range of abduction in scapula plane (scaption) less than 90°. (4) a history of previous shoulder surgery on the affected shoulder. Ultimately, two patients were excluded due to real-time tissue elastography measurement deficiencies, particularly insufficient manual compression pressure applied to the acoustic coupler, and two patients due to a inability to keep their arm passive position at 90° scaption. In total, twenty-four patients (mean age, 66 ± 6 years) were included. The study protocol was approved by the Ethics Committee (Yoshioka Hospital, study no. YHTIB-2021-014). Written informed consent to participate in this study was obtained from all patients. Intraoperative and clinical evaluations Antero-posterior and medio-lateral diameters of the posterosuperior RC tears were measured intra-operatively, with tear size classified based on the DeOrio and Cofield classification [6]. SSC tears associated with posterosuperior tears were classified using the Lafosse classification as follows: type I, partial lesion of superior one-third; type II, complete lesion of superior one-third; type III, complete lesion of superior two-thirds; type IV, complete lesion of tendon but head centered and fatty degeneration classified as less than or equal to stage 3 (indicating as much or less fat than muscle); type V, complete lesion of tendon but eccentric head with coracoid impingement and fatty degeneration classified as more than or equal to stage 3 (indicating as much or more fat than muscle) [24]. Preoperative active range of motion (ROM) in scaption was measured using a goniometer. Real-time tissue (RTE) evaluations Real-time tissue (RTE) evaluations were performed one day before surgery to assess contractility of the deltoid, ISP, and TMin muscles using real-time tissue (RTE) elastography by one examiner (K.H.). A diagnostic ultrasound system (Noblus; Hitachi-Aloka Medical Japan, Tokyo, Japan) with a linear array probe (L-64; Hitachi-Aloka Medical Japan) combined with an acoustic coupler (EZU-TECPL1; Hitachi-Aloka Medical Japan: 22.6 ± 2.2 kPa) was used to obtain the measurements. Briefly, patients were first instructed to sit on a chair with the affected arm placed on a table and hold their arm at 30°, 60°, and 90° of shoulder scaption with a neutral rotation (Fig. 1). The deltoid, ISP, and TMin muscles were scanned to obtain B-mode images, and elastography images were obtained by performing cyclic manual compression with the probe. Ultrasound probe orientation of each muscle was based on previous studies [14–16, 34]. Briefly, to obtain measurements for the deltoid muscle, the probe was placed on the middle region between the scapular acromion and deltoid tuberosity, and rotated until muscle fibers from this muscle could be visualized. The ISP muscle was evaluated by placing the ultrasound probe on the lateral side of muscle and parallel to the direction of the intramuscular tendon and then translated down and medially so that muscle fibers could be visualized. The TMin muscle was imaged by first placing the probe transverse to the muscle at the maximum muscle thickness followed by a rotation of the probe until muscle fibers could be visualized (Fig. 2). Patients were assessed by RTE while at passive and active states, consisting on the patient holding the affected arm against gravity for five seconds. After continuous scanning, three random images at passive and active states were selected to calculate mean ± SD. Elasticity outcomes from the muscles were calculated as the strain ratio (SR) which describes the relative value in reference to the stiffness of the acoustic coupler. In other words, a higher SR indicates that the tissue is softer. Two SR values were defined: SR value at passive state, as an indicator of the muscle elasticity at passive state, and SR value at active state, representing the sum of muscle elasticity at passive state and the elasticity produced by active contraction. The difference between the SRs at passive and active states is defined as muscle activity, reflecting muscle contractility [14–16, 32, 33]. Previous studies on healthy subjects showed activity of the muscles to increase with elevation angles [10, 14, 16]. Based on these observations, activity of the deltoid, ISP and TMin muscles were quantified by defining a muscle activation ratio as the ratio of muscle activity at the initial elevation angle of 30° and the muscle activity at 90°. Measurement position for the real-time tissue elastography (RTE) measurements. A Passive state: patients were instructed to sit on a chair with the affected arm placed on a table. B Active state: patients were instructed to hold their arm at 30°, 60° and 90° of shoulder abduction in the scapula plane with a neutral rotation Real-time tissue elastography (RTE) measurements. (A) Probe orientation: a) middle deltoid (MD), b) infraspinatus (ISP), c) teres minor (TMin) muscles. (B) B-mode and RTE images at passive and active states for one patient Group classification Patients were categorized into normal and abnormal activation groups based on the activation ratio of the deltoid muscle, using the mean activation ratio (0.455 ± 0.168) established in a previous study on healthy subjects [16]. Patients with an activation ratio within two standard deviations (≤ 0.791) were classified into the normal activation group (n = 11), while those with an activation ratio exceeding this range (> 0.791) were classified into the abnormal activation group (n = 13). Muscle volume assessment Given that the deltoid muscle in the abnormal activation group exhibited consistently high activity throughout the entire range of elevation, we explored whether consistently high deltoid muscle activity might be linked to hypertrophy by comparing muscle volumes between normal and abnormal activation groups. MRI was performed a day before surgery to assess the occupation ratio for the deltoid and remaining RC muscles using a 1.5-T scanner (Vantage M-Power Ver.3.0, Canon Medical Systems, Tochigi, Japan). Evaluations were performed using T1-weighted images (repetition time: 450 ms, echo time: 12 ms, field of view (FOV): 16 × 16 cm, slice thickness: 5 mm in the oblique axial plane and 6 mm in oblique sagittal plane). Muscle area measurements were obtained implementing the criteria by Henninger et al. for the deltoid muscle [13] and Kikukawa et al. for the RC muscles [21, 22, 29] (Fig. 3). Deltoid muscle area and humeral head area were tracked at the top of the greater tuberosity on the axial image. SSC, ISP, TMin, and anatomic external rotator (a-ER) muscles area, lateral margin of the scapula, inferior margin of the TMin muscle, and medial margin of the deltoid, were tracked at the most lateral oblique sagittal image in which the scapular spine was in contact with the scapular body. Occupational ratio for each muscle was measured as: deltoid muscle = deltoid muscle area/humeral head area, SSC muscle = SSC muscle area/a-ER muscle area, ISP muscle = ISP muscle area/a-ER muscle area, and TMin muscle = TMin muscle area/a-ER muscle area. Definitions for muscle atrophy and hypertrophy were derived from the 95% confidence interval of the occupational ratios of the ISP and TMin muscles in the intact RC group, as reported in Kikukawa’s study [21, 22, 29]. ISP atrophy was defined as an occupation ratio < 0.586, TMin atrophy as < 0.112, and TMin hypertrophy as > 0.288. MRI images for cross-sectional area measurements of the A deltoid muscle and B remaining rotator cuff muscles. Deltoid muscle area and humeral head area were tracked at the top of the greater tuberosity on the axial image. Subscapularis (SSC), infraspinatus (ISP), teres minor (TMin), and the anatomic external rotator (a-ER) muscles area, lateral margin of the scapula, the inferior margin of the TMin muscle, and the medial margin of the deltoid, were tracked at the most lateral oblique sagittal image in which the scapular spine was in contact with the scapular body. The occupational ratio of the deltoid muscle = deltoid muscle area/humeral head area, the SSC muscle = SSC muscle area/a-ER muscle area, the ISP muscle = ISP muscle area/a-ER muscle area, and the TMin muscle = TMin muscle area/a-ER muscle area Statistical analysis A sample size calculation was performed using G*Power 3.1 software (Heinrich Hein University, Dusseldorf, Germany), based on differences in deltoid muscle activity at 30° scaption between the normal and abnormal deltoid activation groups. Data from three patients per group in our pilot study indicated that 9 patients per group were required to achieve α = 0.05 and a power (1 – β) of 0.80. IBM SPSS Statistics, version 29.0 (IBM Corp., Armonk, N.Y., USA) was used for all statistical analyses. Intraclass correlation coefficient (ICC1,3; one-way random-effects, absolute agreement, multiple measurements) evaluated the intra-rater reliability of the three RTE images obtained from the deltoid, ISP, and TMin muscles for all shoulder positions during passive and active states on all participants [23]. Reliability was classified as poor (< 0.50), moderate (0.50–0.75), good (0.75–0.90), and excellent (> 0.90). Normal distribution of the data was evaluated using Shapiro-Wilk test. Normally distributed data were compared with an independent t-test, and nonnormally distributed data were compared using the Mann-Whitney U test to determine the effect of the demographic factors in the two groups. The χ2 test with Fisher’s exact test was used to compare categorical data in the two groups. A two-factor linear mixed-effects model was implemented to assess differences in muscle activity among the deltoid, ISP and TMin muscles between two groups (normal and abnormal activation groups) and across various elevation angles (30°, 60°, 90°). When significant interactions were observed, a one-way repeated measures ANOVA with Bonferroni-adjusted post-hoc test was used to determine differences in muscle activity at each angle within the groups. Additionally, an independent t-test was conducted to examine differences in muscle activity between two groups across angles. Linear regression analysis was used to investigate independent factors such as age, sex, tear antero-posterior and medio-lateral diameters of the tears, Lafosse classification, active ROM in scaption, and muscle activation ratio of the ISP and TMin muscles related to the activation ratio of the deltoid muscle [30]. Significance was set at p < 0.05. Results Patient demographics Patient demographics, shown in Table 1, were generally similar between the normal and abnormal activation groups. However, a significant difference was observed in the Lafosse classification (p = 0.009), where the abnormal activation group had a higher incidence of SSC tears combined with posterosuperior tears, including six type I, four type II, and one type IV tears. Table 1Characteristics of patients in each deltoid activation group. Bold values indicate statistically significant results (p < 0.05) and their corresponding valuesNormal activation groupAbnormal activation groupp-ValueEffect sizeNumber of patients1113Age, y, mean ± SD65.2 ± 6.667.3 ± 5.80.431Sex, male/female, n5/69/40.408Affected arm side, dominant, n7120.357Posterosuperior tear, mm, mean ± SDAntero-posterior diameter14.9 ± 6.519.5 ± 9.10.082Medio-lateral diameter15.6 ± 7.419.2 ± 9.20.342Cofield classification, n> 0.999Small12Medium99Large12Lafosse classification 0.009 0.666 Type I 0 6 Type II 3 4 Type III 0 0 Type IV 0 1 ROM, scaption, degree, mean ± SD145 ± 33133 ± 210.252 Characteristics of patients in each deltoid activation group. Bold values indicate statistically significant results (p < 0.05) and their corresponding values RTE measurements ICC(1,3) values for RTE measurements were excellent for all shoulder scaption positions at passive state (deltoid muscle: 0.98, 0.98 and 0.98; ISP muscle: 0.97, 0.96 and 0.97; TMin muscle: 0.95, 0.95 and 0.95 for scaption 30°, 60°, and 90°, respectively) and at active state (deltoid muscle: 0.97, 0.95 and 0.94; ISP muscle: 0.94, 0.95 and 0.93; TMin muscle: 0.97, 0.96 and 0.96). Results from a two-factor linear mixed-effects model are shown in Table 2. Significant interactions were observed between deltoid activation groups and elevation angles in the SRs at passive and active states and in the deltoid muscle activity. Significant interactions were also observed in the TMin muscle activity. Additionally, significant main effects of elevation angles were found in the SRs at passive and active states for the TMin muscle. Table 2Statistical results of linear mixed effect model for each muscle. Bold values indicate statistically significant results (p < 0.05) and their corresponding values Main effect InteractionDeltoid activation groupElevation angleMuscleF p Effect sizeF p Effect sizeF p Effect size Strain raito at passive state Deltoid 7.820 0.001 0.262 0.6110.4430.0271.4170.2530.061ISP0.6140.5460.0270.0510.824< 0.0012.6140.0850.106TMin1.0370.3630.0451.0180.3240.044 4.739 0.014 0.177 Strain raito at active state Deltoid 5.598 0.007 0.203 16.163 < 0.001 0.424 40.361 < 0.001 0.647 ISP0.2940.7470.0130.1940.6640.0091.2310.3020.053TMin1.5480.2240.0660.2540.6190.011 8.387 < 0.001 0.276 Muscle activity Deltoid 29.564 < 0.001 0.573 0.7600.3930.033 31.330 < 0.001 0.587 ISP0.0250.976< 0.0010.0070.935< 0.0010.5780.5650.026TMin 4.081 0.024 0.156 0.0770.784< 0.0013.1210.0540.124 Statistical results of linear mixed effect model for each muscle. Bold values indicate statistically significant results (p < 0.05) and their corresponding values At passive states (Fig. 4A), no significant differences in SR were observed between the two groups for each muscle. However, in the abnormal activation group, SR in the deltoid muscle decreased linearly with increasing elevation angles (p < 0.05). Additionally, SR in the TMin muscle showed a significant difference between 60° and 90° (p = 0.012). At active states (Fig. 4B), the deltoid muscle in the abnormal activation group exhibited a significantly lower SR, indicating greater stiffness, compared to the normal activation group at 30° (p = 0.001), 60° (p < 0.001), and 90° (p = 0.044). In the normal activation group, SR decreased linearly with elevation angles, showing significant differences between 30° and 60° (p = 0.009), 30° and 90° (p < 0.001), and between 60° and 90° (p = 0.002). In contrast, the abnormal activation group only showed significant differences between 30° and 90° (p = 0.008). In the ISP muscle, both the normal and abnormal activation groups exhibited relatively constant SR, with no significant differences between the groups and elevation angles. In the TMin muscle, SR in the normal activation group was the lowest, indicating greater stiffness, at 30° and increased linearly as elevation angles to 90° (p < 0.05). On the contrary, SR in the abnormal activation group remained almost constant throughout elevation angles. RTE measurement outcome of the deltoid, infraspinatus (ISP) and teres minor (TMin) muscles in normal and abnormal activation groups. A strain ratio at passive state, B strain ratio at active state, C muscle activity. *significant difference in strain ratio or muscle activity between elevation angles. #significant difference in strain ratio or muscle activity between normal and abnormal activation groups In the muscle activity assessments (Fig. 4C), deltoid muscle activity in the normal activation group increased linearly with elevation angles, with significant differences between 30° and 60° (p = 0.04), 30° and 90° (p < 0.001), and 60° and 90° (p = 0.001). On the other hand, muscle activity in the abnormal group remained almost constant among angles and was significantly higher than that of the normal activation group at 30° scaption (p = 0.037). In the ISP muscle, both the normal and abnormal activation groups exhibited relatively constant activity, with no significant differences between the groups and elevation angles. In the TMin muscle, muscle activity in the normal activation group was highest at 30° and decreased linearly with increasing elevation angle up to 90° (p = 0.032). However, activity in the abnormal activation group remained almost constant among angles. Although activity at 30° was relatively higher in the normal activation group compared to the abnormal activation group, no significant differences were observed between the groups (p = 0.276). Linear regression analysis outcomes on the activation ratio of the deltoid muscle are shown in Table 3. However, none of the clinical factors demonstrated a significant association with the activation ratio of the deltoid muscle. Table 3Linear regression analysis to determine the predictor for activation ratio of the deltoid musclePredictorBβ R 2 p-ValueCofield classification0.1430.2190.0480.304Lafosse classification0.0960.3050.0930.147Activation ratio of the ISP muscle−0.031−0.1020.0040.772Activation ratio of the TMin muscle−0.045−0.2300.0530.280ISP infraspinatus muscle, TMin teres minor muscle Linear regression analysis to determine the predictor for activation ratio of the deltoid muscle ISP infraspinatus muscle, TMin teres minor muscle Assessments for the occupational ratio Occupational ratio outcomes for each muscle and patients exhibiting atrophy of the SSC, ISP and TMin muscles or hypertrophy of the TMin muscle for each group are shown in Table 4. The occupation ratio of the ISP muscle was significantly lower in the abnormal activation group compared to the normal group (p = 0.018), whereas the occupation ratios of the deltoid (p = 0.047) and TMin (p = 0.047) muscles were significantly higher in the abnormal activation group. For the TMin muscle, the abnormal activation group had a significantly larger number of patients with hypertrophy (p = 0.047). There were no significant differences in the occupation ratio of the SSC muscle between the two groups. Table 4Occupation ratio of the deltoid, subscapularis(SSC), infraspinatus (ISP) and teres minor (TMin) muscles. Bold values indicate statistically significant results (p < 0.05) and their corresponding valuesNormal activation group Abnormal activation group p-valueEffect sizeOccupation ratio, Mean (SD)Deltoid1.61 ± 0.311.96 ± 0.380.0470.408SSC1.16 ± 0.171.05 ± 0.240.134 ISP 0.64 ± 0.09 0.09 ± 0.14 0.018 0.479 TMin 0.20 ± 0.07 0.27 ± 0.09 0.047 0.408 Atrophy, n (%)SSC0 (0)0 (0)> 0.999ISP3 (27)6 (46)0.423TMin2 (18)2 (15)0.637Hypertrophy, n (%) TMin 2 (18) 9 (69) 0.047 0.438 Occupation ratio of the deltoid, subscapularis(SSC), infraspinatus (ISP) and teres minor (TMin) muscles. Bold values indicate statistically significant results (p < 0.05) and their corresponding values Discussion This study investigated factors associated with functional differences in the behavior of the deltoid muscle in patients with rotator cuff (RC) tears by comparing tear types and the behavior of the remaining RC muscles between patients with normal and abnormal deltoid activation. We hypothesized that the functional difference of the deltoid muscle would be influenced by tear size and compensatory activity of the remaining RC muscles. Our results showed that patients with abnormal deltoid activation, i.e.: abnormal activation group, had a higher incidence of subscapularis (SSC) tears combined with posterosuperior RC tears, and patients with normal deltoid activation, i.e.: normal activation group, exhibited high compensatory activity of the teres minor (TMin) muscle in the initial range of elevation. Furthermore, the abnormal activation group presented a larger occupation ratio of the deltoid and TMin muscles compared to the normal activation group. Previous clinical studies have reported increased deltoid muscle activation as a compensatory mechanism for the functional loss of the RC muscles in patients with RC tears [5, 12] and following RC repair [17, 26, 30]. The current study focuses on spatial changes in deltoid muscle behavior during arm elevation and to identify two distinct patterns of deltoid muscle behavior in patients with RC tears. In the normal activation group, the deltoid muscle maintained a constant stiffness at passive and active states, stiffness progressively increased with elevation angles, reaching its peak at 90° of scaption. This trend mirrors observations in healthy individuals without RC tears [16]. In contrast, the abnormal activation group showed increasing stiffness with increasing elevation angles at passive state and significantly greater stiffness at active state across all elevation angles compared to the normal activation group. The abnormal activation group had a significantly higher number of patients with more than one-third of the SSC muscle torn, combined with posterosuperior rotator cuff tears, compared to the normal activation group. In a biomechanical cadaveric study, Yoo et al. demonstrated that SSC tears affecting more than half of the tendon, when combined with posterosuperior RC tears, lead to significant superior and posterior instability of the humeral head [31]. Hansen et al. showed that increased force production by the deltoid and TMin muscles is required in the early phase of abduction to prevent significant superior migration of the humeral head in massive RC tears combined with an SSC tear [11]. These findings are consistent with our clinical observations and suggest that the presence of SSC tears combined with posterosuperior RC tears may contribute to compensatory activation of the deltoid muscle during early arm elevation. However, linear regression analysis revealed no significant association between the Lafosse classification and the deltoid muscle activation ratio outcomes. Further investigation will be required to elucidate the relationship between SSC tears in large and massive RC tears and compensatory activation of the deltoid muscle during arm elevation. We observed a high number of patients (6/13) exhibiting the shrug sign [18] in the abnormal activation group when compared to the normal group (0/11). Burkhart et al. emphasized that patients with extensive tears extending from the anterior to posterior regions of the RC may exhibit the ‘shrug sign’, a clinical manifestation resulting from the disruption of the transverse force couple between the anterior and posterior RC muscles, leading to an inability to maintain a stable fulcrum during active arm elevation [4]. Kelly et al. reported increased upper trapezius activity, contributing to scapular upward rotation during arm elevation, in patients with symptomatic RC tears [19]. Therefore, the presence of SSC tears, in conjunction with posterosuperior RC tears, could disrupt the transverse force couple, leading to compensatory activation of other shoulder girdle muscles, such as the deltoid, potentially resulting in a positive shrug sign. Some clinical studies have implemented ultrasound elastography to investigate the compensatory strategies of the deltoid and RC muscles during arm elevation following RC repair [17, 26, 30]. Ishikawa et al. measured the muscle activity of the supraspinatus (SSP) and deltoid muscles in patients following RC repair, demonstrating that recovery of the SSP muscle activity led to decreased compensatory activity of the deltoid muscle [17]. In our study, we found that the compensatory activation of the deltoid muscle may be associated with the compensatory activation of the TMin muscle in patients with RC tears. Previous studies have shown that activation of the TMin muscle in healthy volunteers increases with elevation angle, peaking from 120° to 135° of abduction [10]. In contrast, in the normal activation group of our study, TMin muscle activity peaked at 30° scaption and decreased linearly to 90°, representing a contrasting trend compared to healthy volunteers. Alpert et al. demonstrated that the activation peak of the TMin muscle shifted to the initial phase with increasing loads during scaption [1]. These findings suggest that the activation peak may have shifted to the initial range of motion as a compensatory behavior for the functional deficit of the torn RC muscles in the normal activation group. In the abnormal activation group, TMin muscle activity remained consistent across all angles, and activity at 30°, while not significantly, was lower than that of the normal activation group. These findings suggest that the relatively greater activity of the TMin muscle at 30° in the normal activation group may be associated with lower activity of the deltoid muscle in the initial range of elevation. Therefore, compensatory activity of the TMin muscle in the initial range of motion may reflect a compensatory contribution of the remaining posterior RC muscles. In this study, we hypothesized that changes in muscle activity of the deltoid and remaining RC muscles might also affect changes in muscle volume. Our results showed that in the abnormal activation group, the occupation ratios of the deltoid and TMin muscles were significantly larger, while the ISP muscle was significantly smaller compared to the normal activation group. Upon examining factors related to deltoid hypertrophy, we found that the deltoid activation ratio and the occupation ratio of the TMin muscle were significantly associated with deltoid muscle hypertrophy. Previous studies have demonstrated that TMin hypertrophy occurs in response to ISP atrophy [21], and our findings similarly revealed a significant negative correlation (r = −0.629, p < 0.001) between the occupation ratios of the ISP and TMin muscles (Fig. 5A). Furthermore, our results showed a significant positive correlation (r = 0.563, p = 0.004) between the occupation ratios of the deltoid and TMin muscles (Fig. 5B). Previous clinical studies reported that patients showing hypertrophy of the TMin muscle restored shoulder strength and range of external rotation [22] and mitigated superior migration of the humeral head in posterosuperior RC tears [29]. In this study, we observed hypertrophy of the TMin muscle in 10 out of the 24 cases, with most of the patients (8 out of 10) belonging to the abnormal activation group. Both the deltoid and TMin muscles in the abnormal activation group exhibited relatively higher stiffness at passive and active states, compared to the normal group, and both muscles maintained consistent activity across all elevation angles. This suggests that these muscles were continuously active throughout the elevation motion, which may have contributed to these muscle hypertrophy. Correlation A between occupation ratio of the infraspinatus (ISP) and teres minor (TMin) muscles B between the occupation ratio of the deltoid and TMin muscles. Green shaded area (occupation ratio < 0.586) represents patients with atrophy of ISP muscle. Blue shaded area (occupation ratio < 0.112) represents patients with atrophy of TMin muscle, while the red shaded area (occupation ratio > 0.288) represents patients with hypertrophy of TMin muscle This study has several limitations. Firstly, due to the anatomical structure, it was not possible to measure SSC muscle stiffness using real-time tissue elastography (RTE). Instead, we assessed the SSC muscle based on intraoperative findings and muscle volume. Additional research is required to gain a comprehensive understanding of SSC muscle behavior in patients with various RC tear conditions. Secondly, the interpretation of ISP activity requires caution. Although RTE measurements in this study assessed the contractile behavior of the ISP muscle, the extent of ISP tendon involvement in posterosuperior RC tears may have varied among patients. Therefore, the extent of ISP tendon involvement may have contributed to variability in ISP activity, and it remains unclear to what extent the measured ISP muscle activity was transmitted through the residual tendon and contributed to humeral head stability. Thirdly, it is important to note that the number of patients with more extensive SSC tears (more than two-thirds involvement) was limited, with most patients having one-third SSC tears in the abnormal activation group. Lee et al. reported a high failure rate in massive repairs of posterosuperior RC tears that extended beyond the superior two-thirds of the subscapularis tendon [25]. Further investigations on the behavior of deltoid and RC muscles will be necessary to fully understand the influence of SSC tear size on the remaining muscles. Lastly, this study only included patients who could actively achieve more than 90° of scaption. Our findings are particularly relevant to patients with RC tears who maintain a higher range of active motion and may not directly apply to individuals with severe shoulder dysfunctions, such as pseudoparalysis. Despite this methodological decision which can be considered a study limitation, the observed changes in the behavior of the deltoid and TMin muscles provide valuable insights into the adaptive changes in shoulder dynamics in response to RC tear progression, improving our understanding of compensatory mechanisms in patients with RC tears. In conclusion, our results suggest the presence of SSC tears combined with posterosuperior RC tears may be associated with compensatory activation of the deltoid muscle in the early phase of arm elevation. In addition, an increase in TMin activity during the initial range of elevation may reflect compensatory contribution within the remaining posterior RC muscles. The abnormal activation group also exhibited larger occupation ratios of the deltoid and TMin muscles compared with the normal activation group, and deltoid muscle hypertrophy was associated with TMin muscle hypertrophy. Acknowledgements We would like to thank the head of the Yoshioka Hospital, Dr. Shinya Yoshioka, and all of the staff department of orthopaedic surgery at Yoshioka hospital, whose support made this project possible. Author contributions K.H. (Funding acquisition; Conceptualization; Data curation; Formal analysis; Investigation; Methodology; Project administration; Writing—original draft; Writing—review and editing). R.O. (Data curation, Writing—review and editing). T.Y. (Methodology; Writing—review and editing). T.U.(Data curation, Writing—review and editing). J.N.(Data curation, Writing—review and editing). H.G. (Writing—review and editing). N.M. (Conceptualization; Data curation; Methodology; Project administration; Supervision; Writing —review and editing) Funding One of the authors (KH) has received research funding from JSPS (Grant number 22J22344). Data availability Sequence data that support the findings of this study have been stored at the Graduate School of Health Sciences, Yamagata Prefectural University of Health Sciences, Yamagata, Japan. Declarations Conflict of interest The authors declare no competing interests. Footnotes Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. References Associated Data Data Availability Statement Sequence data that support the findings of this study have been stored at the Graduate School of Health Sciences, Yamagata Prefectural University of Health Sciences, Yamagata, Japan.