Physical Therapy Korea

Rotator cuff tears are a common cause of shoulder pain that can lead to impaired shoulder joint function and limit daily activities [1]. Although conservative treatment is commonly favored as the initial approach, it is essential to determine the appropriate timing for surgical intervention is critical [2]. Typically, conservative treatment is recommended for small-to medium-sized tears, whereas surgery is required for large or massive tears require surgery [3]. Surgery is recommended when no improvement is observed in symptoms such as pain, range of motion (ROM), and inflammation after approximately 6 months of conservative treatment [3]. However, despite considerable advancements in surgical techniques, retear rates remain high (11%–57%) [4,5], and delayed primary surgery has been linked to an increased likelihood of requiring revision surgery [6].

A recent consensus on rotator cuff tears suggests that surgery is not recommended for large or massive tears in the absence of symptoms or functional impairments [7,8]. Furthermore, the anatomical characteristics of rotator cuff tears may not be a determining factor in the decision-making process for surgical intervention [9]. However, it is widely recognized that rotator cuff tears disrupt the biomechanical synergy between the rotator cuff and the deltoid, leading to decreased shoulder muscle strength and function [10].

Muscle weakness is commonly observed in the shoulder abductors and external rotators [11,12]. Functional deficits arise from a combination of muscle weakness, pain, and a limited ROM [13]. Various methods are available to assess these weaknesses and functional declines. Isokinetic muscle strength testing is often employed to measure muscle weakness [11,12], while the American Shoulder and Elbow Surgeons (ASES) and Constant-Murley shoulder scores are commonly used to evaluate function [14,15]. ASES is a reliable, valid, and responsive outcome measure [14]. Similarly, the Constant score is considered an excellent tool because the direction of measurement in forward flexion and scapular abduction does not affect strength, pain intensity, or total score [15].

Surgical intervention for rotator cuff tears is more likely to be considered when the tear size is larger, the symptoms last longer, and the functional comorbidity index is higher at the initial clinical presentation [2]. Extensive studies have been conducted on muscle recovery and functional improvement following surgery [16-19], but few studies have compared preoperative muscle strength and function. Furthermore, while previous research has examined preoperative indicators, most have focused on morphological aspects, such as differences in ROM, tear size, and fatty infiltration on magnetic resonance imaging, rather than shoulder muscle strength and function [20]. The latter recognizes the importance of strength and function after surgery; such studies have examined preoperative ROM or abnormalities in radiographic imaging. However, there is a notable lack of research focusing on both groups’ preoperative shoulder muscle conditions and functional outcomes.

Understanding the preoperative condition of the shoulder is important because it significantly affects surgical outcomes [2,20,21]. Muscle strength should not be considered in general terms but in more specific measurements. Various methods can evaluate muscles, such as peak muscle torque relative to body weight, deficits, and muscle ratios. Shoulder function questionnaires are also diverse and are tailored for specific purposes. Therefore, conducting a comprehensive comparison of these factors is essential to clarify the preoperative variables that differ between the surgical and conservative treatment groups. The purpose of this study is to compare preoperative shoulder strength and function between the surgical and conservative treatment groups and to identify trends in those who did not undergo rotator cuff repair. This study hypothesized that the two groups would have significant differences in shoulder muscle strength and function.

1. Patients

This study retrospectively reviewed the medical records of 404 patients who presented with rotator cuff tears between January 2008 and December 2016. Although patients were recruited nationwide, most resided in Seoul, and the participants were randomly selected. After excluding 89 patients with missing data, 315 patients were included in the study. The inclusion criteria were as follows: (1) patients with supraspinatus tears, (2) patients capable of performing isokinetic muscle strength tests, and (3) patients over 20 years old. The exclusion criteria were as follows: (1) patients with isolated subscapularis tears, (2) those who underwent revision surgery, and (3) those covered under industrial accident insurance. A flowchart of the patient selection process is shown in Figure 1. Among the included patients, 145 underwent surgery, and 170 received conservative treatment.

Figure 1. Flowchart of study enrollment.

A pilot study was conducted prior to this study. Using G*Power 3.1 software, we calculated the required sample size to be 65, based on an effect size of 0.78 standard deviations, a significance level of 0.05, and a statistical power of 0.80.

This study was approved by the Ethics Committee of Yonsei University Mirae campus (IRB no. 1041849-202411-BM-233-01). It retrospectively reviewed the medical records to gather the demographic and clinical characteristics of the enrolled patients.

2. Outcome

1) Isokinetic muscle strength test

Following the guidelines provided by Biodex Corp., isokinetic muscle strength for each testing posture was assessed using the Biodex System 3 (Biodex Corp.). Shoulder external and internal rotation tests were also performed in the seated position, ranging from –45° to 75° (Figure 2), while shoulder abduction and adduction tests were conducted in the seated position, ranging from 0° to 90° (Figure 3). Peak torque and peak torque per body weight (PT/BW) were assessed through four repetitions of abduction, adduction, external rotation, and internal rotation of both shoulder joints at a speed of 60 °/s [22]. The system automatically recorded the highest recorded data from the four repetitions of peak torque and PT/BW. The deficit value was determined to assess how much weaker the involved side strength was compared to the uninvolved side strength. Deficits were determined as percentages using the following formula: Deficit% = (1 – strength of affected side/strength of uninvolved side) × 100. The test-retest reliability of muscle strength assessments conducted with the isokinetic dynamometer was deemed excellent, with reliability coefficients ranging from 0.95 to 0.99 [23].

Figure 2. Isokinetic muscle strength test for internal and external rotator.

Figure 3. Isokinetic muscle strength test for abductor and adductor.

2) Shoulder function scores

(1) American Shoulder and Elbow Surgeons score

The ASES score is widely used for evaluating shoulder joint function during activities of daily living [24]. This 10-item questionnaire assesses shoulder pain and function, with a total score of 50% for pain and 50% for function, resulting in a maximum score of 100. Higher scores indicate a more favorable subjective assessment of the shoulder condition. The ASES score is advantageous owing to its high reproducibility, ease of application in clinical settings, and applicability to a wide range of shoulder conditions.

(2) Constant-Murley shoulder score

The Constant-Murley score is a 100-point scoring system divided into four subscales: pain (15 points), activities of daily living (20 points), ROM (40 points), and strength (25 points). The pain and activities of daily living subscales are self-reported by patients. The original version graded pain scores as follows: none = 15, mild = 10, moderate = 5, and severe = 0. Since then, pain scoring has been replaced by the visual analog scale [25].

3. Procedure

Shoulder isokinetic muscle strength tests and two functional scores were performed on all patients on their first outpatient visit. After registering for outpatient care, patients visited the sports medical center in the hospital for testing. The results were subsequently reviewed during their outpatient appointment. The surgical group underwent retesting on the morning of surgery. The data from the conservative treatment group from the first outpatient visit were compared with the surgical group retest data on the morning of surgery. The conservative treatment group received injections or rehabilitation as needed following their outpatient visit. Rehabilitation was conducted sequentially, beginning with initial ROM recovery, then muscle strength recovery, and finally functional recovery.

4. Statistical Analysis

All statistical analyses were performed using IBM SPSS software (ver. 22.0 for Windows; IBM Co.). The Kolmogorov–Smirnov normality test was used to assess the normality of the data. To identify the patient characteristics, sex, age, height, weight, and body mass index were calculated as averages and standard deviations. An independent t-test assessed differences in shoulder muscle strength and functional scores between the rotator cuff repair and conservative treatment groups. Statistical significance was set at p < 0.05.

The study included 315 participants, comprising 190 males (approximately 60%) and 125 females (approximately 40%) across both groups. The mean age of participants was 53.3 ± 9.9 years, with an average height of 164.6 ± 8.6 cm, weight of 66.0 ± 11.5 kg, and body mass index of 24.3 ± 4.2 kg/m². Baseline characteristics of the surgery and conservative treatment groups are shown in Table 1. No significant differences were observed between the two groups.

Table 1 . Baseline characteristics of patients with and without arthroscopic rotator cuff repair (N = 315).

Variable	Surgery group (n = 145)	Non-surgery group (n = 170)	p-value
Sex (Male/female)	91 (63)/54 (37)	99 (58)/71 (42)	None
Age (y)	56.0 ± 9.3	50.9 ± 9.9	0.075
Height (cm)	163.5 ± 9.1	165.5 ± 8.1	0.142
Weight (kg)	67.1 ± 12.5	65.1 ± 10.4	0.228
Body mass index (kg/m2)	23.7 ± 4.8	24.9 ± 5.6	0.064

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

In terms of shoulder strength deficits, the conservative treatment group demonstrated lower deficits in both the external and internal rotators than the surgery group. The external-to-internal rotator ratio was also higher in the conservative treatment group (Table 2). However, no significant differences were found between the groups in peak torque and PT/BW for either the external or internal rotators. Likewise, shoulder abductor and adductor deficits were lower in the conservative treatment group than in the surgery group (Table 3); however, the peak torque, PT/BW, and abductor per adductor ratio were not significantly different.

Table 2 . Comparison of isokinetic muscle strength test for external and internal rotator between the two groups (N = 315).

	Surgery group (n = 145)	Non-surgery group (n = 170)	p-value
External rotator peak torque	12.1 ± 7.3	15.1 ± 7.3	0.539
External rotator PT/BW (%)	17.7 ± 9.7	22.7 ± 9.2	0.984
External rotator deficit (%)	26.3 ± 33.8	11.3 ± 23.9	0.001*
Internal rotator peak torque	27.6 ± 11.7	29.6 ± 11.6	0.474
Internal rotator PT/BW (%)	41.1 ± 16.8	45.2 ± 14.6	0.083
Internal rotator deficit (%)	17.1 ± 26.1	11.7 ± 15.5	0.001*
External/internal rotator ratio	44.2 ± 19.6	50.9 ± 16.8	0.003*

Values are presented as mean ± standard deviation. PT/BW, peak torque per body weight. *The mean difference is significant at the 0.05 level..

Table 3 . Comparison of isokinetic muscle strength test for abductor and adductor between the two groups (N = 315).

	Surgery group (n = 145)	Non-surgery group (n = 170)	p-value
Abductor peak torque	19.3 ± 11.3	24.4 ± 18.7	0.269
Abductor PT/BW (%)	28.2 ± 15.3	34.7 ± 15.9	0.959
Abductor deficit (%)	30.7 ± 31.6	18.2 ± 25.3	0.036*
Adductor peak torque	45.8 ± 21.9	51.2 ± 19.3	0.127
Adductor PT/BW (%)	67.7 ± 31.2	81.0 ± 60.9	0.986
Adductor deficit (%)	21.9 ± 28.4	9.8 ± 16.8	0.001*
Abductor/adductor ratio	41.5 ± 23.0	46.8 ± 23.0	0.984

Values are presented as mean ± standard deviation. PT/BW, peak torque per body weight. *The mean difference is significant at the 0.05 level..

Functional scores showed some differences between the groups; the conservative treatment group had a higher average ASES score than the surgical group (Table 4). However, the two groups had no difference in the Constant-Murley shoulder scores.

Table 4 . Comparison of shoulder function scores between the two groups (N = 315).

Score	Surgery group (n = 145)	Non-surgery group (n = 170)	p-value
ASES	57.5 ± 16.2	74.5 ± 14.8	0.047*
Constant-Murley	64.3 ± 9.6	67.4 ± 10.3	0.980

Values are presented as mean ± standard deviation. ASES, American Shoulder and Elbow Surgeons. *The mean difference is significant at the 0.05 level..

This study examined differences in shoulder muscle strength—specifically, peak torque, PT/BW, deficits, and muscle balance ratios—and functional scores between rotator cuff surgery and conservative treatment groups. The results indicated that the conservative treatment group exhibited lower average deficits in shoulder muscle strength across the external rotator, internal rotator, abductor, and adductor muscles than the surgery group. Additionally, the external-to-internal rotator ratio was higher in the non-muscle surgery group. In terms of functional outcomes, the non-surgical group achieved a higher average ASES score than the surgical group.

In the present study, the deficits in all four isokinetic muscle strength tests (external rotation, internal rotation, abduction, and adduction) were lower in the conservative treatment group. This finding suggests that shoulder strength may be an important indicator in preoperative assessments for determining the need for surgery. However, it is necessary to consider why only the deficit, rather than the peak torque or PT/BW, differed among the shoulder strength indicators. The isokinetic muscle strength test provides a range of data including peak torque, PT/BW, average power per repetition, total work done, side-to-side deficit, and muscle ratio. While previous studies have primarily focused on peak torque and average power [26-28], this study identified a difference specifically in deficits rather than in peak torque or PT/BW. This suggests that, given individual variability in strength, the side-to-side muscle deficit may be more significant than absolute strength itself. Several previous studies also used deficit as a key indicator [11,29], suggesting that it could be a more prominent factor in studies comparing or analyzing muscle strength.

The external to internal rotator strength ratio was higher in the non-surgical group, indicating relatively greater external rotator strength. The importance of the external rotators in shoulder stability in patients with rotator cuff injuries and shoulder pain has been emphasized in several previous studies [12,30], and the present findings align with this. In the current study, the external per internal rotator ratio was 44.0 ± 19.6 for the surgery group and 50.9 ± 16.8 for the conservative treatment group, whereas Berckmans et al. [31] reported a range of 0.46 to 0.79 in a systematic review. This trend was slightly higher than the results of the current study, and subjects with higher levels of daily activity tended to exhibit higher rotator ratios. In addition, no significant differences were observed in the abductor-to-adductor ratio. However, shoulder abductor muscles are important for joint stability. Previous studies have shown that larger rotator cuff tears are associated with greater shoulder abductor deficits [10,11], whereas smaller deficits are observed in asymptomatic rotator cuff tears [29]. In the present study, the abductor per adductor ratio was 41.5 ± 23.0 for the surgery group and 46.8 ± 23.0 for the conservative treatment group, reflecting a trend consistent with the 0.45 ± 0.14 reported in previous study [32]. This is likely because external and internal rotators are crucial for shoulder stability [31,32]. Therefore, only the external rotator ratio differed significantly between the two groups.

The conservative treatment group achieved a higher average ASES score than the surgery group; however, there was no difference in the Constant-Murley shoulder score between the two groups. The ASES is a shoulder questionnaire that focuses on pain and function, and is commonly used to help determine the need for surgery. Kweon et al. [2] reported that the ASES score is one of the primary questionnaires for surgical decision-making, along with the Western Ontario Rotator Cuff Index and the Veterans Rand 12-Item Health Survey. Similarly, Weekes et al. [13] identified that limitations in shoulder function, as assessed using the ASES score, were the most influential factors prompting patients to undergo rotator cuff repair. Although the Constant-Murley shoulder score is also widely used, it allocates fewer points to pain and function (15 and 20 points, respectively) relative to ROM (40 points) and strength (25 points), making pain and function less weighted than the ASES score. Consequently, only ASES scores showed significant differences between the two groups. These findings suggest that pain and function may be key factors influencing surgical decisions in rotator cuff repair.

This study had some limitations. First, the activity levels of patients were not considered. Although it should have been, strength and functional scores could vary according to activity level. Second, age was not restricted. In the present study, although there were no age differences between the two groups, age strongly affected shoulder strength and function. Third, the retrospective study design may have introduced a selection bias. Fourth, there may be differences in the initial homogeneity of the two groups in clinical practice. Fifth, this study selected subjects treated between January 2008 and December 2016. It is possible that the timing of the subjects’ onset differed significantly.

The conservative treatment group had lower shoulder strength deficits, better muscle balance, and higher ASES scores than the rotator cuff repair group, indicating superior functional outcomes. These findings suggest that assessing preoperative muscle strength and function may provide valuable insights into treatment planning for patients with rotator cuff tears.

None.

None to declare.

No potential conflicts of interest relevant to this article are reported.

Conceptualization: JK, OK. Data curation: JK, IA. Formal analysis: UH. Investigation: JK. Methodology: JK, UH. Project administration: UH. Supervision: UH, OK. Visualization: JK, IA. Writing - original draft: JK. Writing - review & editing: UH, OK.