Clinical application of ultrasound-guided magnetic resonance arthrography for the diagnosis of supraspinatus tendon tears

Introduction

Located at the top of the shoulder cuff, the supraspinatus tendon is in close proximity to the acromion and coracoacromial arch and has a large range of motion. It is prone to friction and impingement, making it susceptible to degeneration and injury (1,2). As the most common type of rotator cuff injury, supraspinatus tendon tear has an incidence of 25% in people over 50 years old and up to 50% in people over 70 years old (1,2). It includes the partial tear and full-thickness tear types, with partial tear further divided into articular-sided, bursal-sided, and intratendinous tear (3). Damage to the supraspinatus tendon often causes shoulder joint pain, which is easily mistaken for shoulder periarthritis by patients. Subsequently, they may engage in activities such as shoulder stretching exercises on their own, worsening the injury, and thus accurate diagnosis is crucial. Musculoskeletal ultrasound (MSKUS) and magnetic resonance imaging (MRI) are important diagnostic methods for determining the presence and subtypes of tear. However, these two methods can produce false positives and false negatives, affecting the clinical diagnosis and treatment strategies (4,5). Magnetic resonance arthrography (MRA) has high reliability in the diagnosis of supraspinatus tendon tear and its subtypes, but it requires injection of gadolinium contrast agent into the glenohumeral joint. However, traditional blind puncture is difficult to operate and carries high risk, resulting in a low success rate, which limits its clinical application (6-8). Furthermore, traditional MRA examination has high diagnostic value only for articular-sided tear when the glenohumeral joint is injected with contrast agent, and its diagnostic value for bursal-sided tear is comparable to that of conventional MRI.

Ultrasound-guided shoulder joint puncture is accurate, safe, and convenient. It can puncture not only the glenohumeral joint but also the subacromial bursa, with a high success rate (9,10). Therefore, combining the ultrasound-guided puncture technique with MRA can leverage each modalities respective advantages to improve the diagnostic accuracy in supraspinatus tendon tear and its subtypes, which may offer considerable clinical benefit. Only a few isolated studies have examined ultrasound-guided MRA (USMRA) of the glenohumeral joint and subacromial bursa with gadolinium contrast agent. Thus, this study aimed to assess the clinical value of USMRA in the diagnosis of supraspinatus tendon tears and its subtypes. We present this article in accordance with the STARD reporting checklist (available at https://qims.amegroups.com/article/view/10.21037/qims-24-765/rc).

Methods

Study participants

This retrospective study was carried out in the Department of Orthopedics at Huzhou Central Hospital. A total of 207 patients with shoulder pain were included from September 2020 to September 2022. This study was conducted in accordance with the Declaration of Helsinki (as revised in 2013) and was approved by the Medical Ethics Committee of Huzhou Central Hospital (No. 202206043-01). Informed consent was provided by all patients. MSKUS and USMRA were performed when patients were suspected of having supraspinatus tendon tears after their medical history and physical examination were evaluated. For those who had concerns regarding ultrasound-guided puncture due to the injection of gadolinium contrast agent into the glenohumeral joint, MSKUS and MRI were conducted. All patients ultimately underwent shoulder arthroscopy.

The inclusion criteria were as follows (11-13): (I) history of shoulder joint pain with limited mobility; (II) suspected supraspinatus tendon tear based on clinical history and physical evaluation; and (III) MSKUS, MRI, or USMRA suggestive of supraspinatus tendon tear. Meanwhile, the exclusion criteria were as follows: (I) presence of infection in the shoulder joint or other body parts; (II) allergy to gadolinium contrast agents; (III) incomplete data of shoulder arthroscopy; (IV) abnormal coagulation function or discontinuation of antiplatelet drugs for less than 3 days; (V) severe heart, liver, kidney, or other organ dysfunction or inability to tolerate the examination; (VI) previous history of congenital deformities, trauma, or surgery of the shoulder joint; and (VII) other rheumatic autoimmune diseases such as rheumatoid arthritis.

Examination methods

MSKUS examination was performed with a DC-80S and R7T ultrasound diagnostic device (Mindray, Shenzhen, China) with a probe frequency of 12 MHz. MSKUS conditions were used, and color Doppler conditions were set to low-pass filtering with a pulse repetition frequency of 400–1,800 Hz. The maximum gain was adjusted to prevent Doppler signals from appearing behind the bone cortex.

MRI examination was performed using a 3-T MRI scanner (GE HealthCare, Chicago, IL, USA) with a shoulder joint coil. The routine scanning planes included transverse, oblique sagittal, and oblique coronal planes. The scanning sequences included spin-echo (SE) T1-weighted imaging (T1WI) in the coronal and oblique sagittal positions [repetition time (TR) 500 ms, time to echo (TE) 22 ms, matrix 256×256, slice thickness 3 mm, interval 0.3 mm], fast SE (FSE) T2WI in the coronal and transverse positions (TR 3,000 ms, TE 85 ms, matrix 256×256, slice thickness 3 mm, interval 0.3 mm), and proton density-weighted imaging (PDWI) in the oblique sagittal position.

For USMRA, the gadolinium contrast agent composition (14) included 0.3 mL of gadopentetic acid dimeglumine, 10 mL of 0.75% ropivacaine, and 10 mL of saline mixed to a total volume of 20 mL. During the USMRA examination of the shoulder joint, preliminary MSKUS examination was performed to determine the presence and quantity of subacromial bursa effusion. Subsequently, a guided puncture of the glenohumeral joint was performed under ultrasound guidance, with 15 mL of gadolinium contrast agent being injected and the subacromial bursa being immediately examined with ultrasound to assess the change in the amount of fluid. Subsequently, a guided puncture of the subacromial bursa was performed, with 5 mL of gadolinium contrast agent being injected, and then the patient was advised to move the shoulder joint appropriately. The MRI examination of the shoulder joint was operated half an hour later with a SE-T1WI fat-suppressed (FS) sequence (TR/TE =565/13 ms), with the remaining scanning parameters and methods being identical to those of the MRI examination of the shoulder joint.

Image interpretation

Image interpretation was jointly carried out by two associate chief physicians specializing in musculoskeletal radiology. When any discrepancy arose during the diagnosis, a consensus was reached through consultation, and in case of continued disagreement, the patient was excluded.

Shoulder arthroscopy

Shoulder arthroscopy was performed using an arthroscopy system (Karl Storz, Tuttlingen, Germany) by one associate chief of joint surgery and attending assistant physicians. The severity of supraspinatus tendon tear was categorized according to the Ellman classification as follows (3): grade I (mild), year depth <3 mm; grade II (moderate), tear depth ≥3 and ≤6 mm; and grade III (severe), tear depth >6 mm.

Statistical analysis

SPSS 21 statistical software (IBM Corp., Armonk, NY, USA) was used for the analyses. Qualitative data were expressed as a rate, quantitative data consistent with a normal distribution were expressed as the mean ± standard deviation, and skewed data are expressed as the median (M) and interquartile range (Q). Group differences were compared with the t-test, χ² test, or Fisher exact test. The McNemar test was used to analyze the diagnostic sensitivity, specificity, positive predictive value, negative predictive value, and accuracy. A P value <0.05 was considered statistically significant.

Results

Basic clinical characteristics

Among the 207 patients, there were 56 males (27.05%) and 151 females (72.95%). The average age of patients was 51.52±10.35 years, with a disease duration of 6.31±2.32 months. The visual analogue scale (VAS) score of preoperative pain was 7.26±3.63, the active range of motion (AROM) score of the shoulder joint was 17.54±6.81, and the Constant-Murley score of the shoulder joint was 45.71±5.21.

In the MSKUS group, arthroscopic examination in 207 patients showed that supraspinatus tendons were torn in 155 cases (155/207, 74.88%), while there were no tears in 52 cases (52/207, 25.12%). The distribution of the subtypes of the supraspinatus tendon tears were as follows: articular-sided tears in 35 cases (35/207, 16.91%), bursal-sided tears in 32 cases (32/207, 15.46%), intratendinous tears in 38 cases (38/207, 18.36%), and full-thickness tears in 50 cases (50/207, 24.15%) (Figure 1). In terms of tear grades, 22 cases had grade I tears (22/155, 14.29%), 39 cases had grade II tears (39/155, 25.16%), and 94 cases had grade III tears (94/155, 61.04%).

Figure 1 The different subtypes of supraspinatus tendon tears on MSKUS. (A) Bursal-sided tears; (B) articular-sided tears; (C) intratendinous tears; (D) full-thickness tears. The arrows indicate the location of the supraspinatus tendon tear. MSKUS, musculoskeletal ultrasound.

In the MRI group, arthroscopic examination in 103 patients revealed that supraspinatus tendons were torn in 77 cases (77/103, 74.76%), while 26 cases had no tears (26/103, 25.24%). The distribution of the subtypes of the supraspinatus tendon tears were as follows: articular-sided tears in 14 cases (14/103, 13.59%), bursal-sided tears in 22 cases (22/103, 21.36%), intratendinous tears in 20 cases (20/103, 19.42%), and full-thickness tears in 21 cases (21/103, 20.39%). In terms of tear grades, 10 cases had grade I tears (10/103, 9.71%), 23 cases had grade II tears (23/103, 22.33%), and 44 cases had grade III tears (44/103, 42.72%).

In the USMRA group, arthroscopic examination in 104 patients indicated supraspinatus tendons were torn in 78 cases (78/104, 75.00%), while there were no tears in 26 cases (26/104, 25.00%). The distribution of subtypes of supraspinatus tendon tears were as follows: articular-sided tears in 20 cases (20/104, 19.23%), bursal-sided tears in 13 cases (13/104, 12.50%), intratendinous tears in 20 cases (20/104, 19.23%), and full-thickness tears in 25 cases (25/104, 24.04%) (Figure 2). In terms of tear grades, 12 cases had grade I tears (12/104, 11.54%), 16 cases had grade II tears (16/104, 15.38%), and 50 cases had grade III tears (50/104, 48.08%).

Figure 2 The different subtypes of supraspinatus tendon tears on USMRA. (A) Bursal-sided tears; (B) articular-sided tears; (C) intratendinous tears; (D) full-thickness tears. The arrows indicate the location of the supraspinatus tendon tear. USMRA, ultrasound-guided magnetic resonance arthrography.

According to the results of arthroscopic examination, there were no significant differences in the presence of supraspinatus tendon tears (χ²=0.05; P=0.97), tear subtypes (χ²=4.48; P=0.81), and tear grades (χ²=1.82; P=0.77) between the MSKUS, MRI, and USMRA groups.

Presence and subtypes of supraspinatus tendon tears in the MSKUS, MRI, and USMRA groups

In the MSKUS group, MSKUS diagnosis showed that supraspinatus tendons in 147 cases were torn (147/207, 71.01%), while 60 cases had no tears (60/207, 28.99%). The distribution of tear subtypes was as follows: articular-sided tears in 37 cases (37/207, 17.87%), bursal-sided tears in 32 cases (32/207, 15.46%), intratendinous tears in 42 cases (42/207, 20.29%), and full-thickness tears in 36 cases (36/207, 17.39%).

In the MRI group, MRI diagnosis showed that the supraspinatus tendons in 76 cases were torn (76/103, 73.79%), while 27 cases had no tears (27/103, 26.21%). The distribution of tear subtypes was as follows: articular-sided tears in 14 cases (14/103, 13.59%), bursal-sided tears in 22 cases (22/103, 21.36%), intratendinous tears in 20 cases (20/103, 19.42%), and full-thickness tears in 20 cases (20/103, 19.42%).

In the USMRA group, the results of USMRA examination showed supraspinatus tendons were torn in 78 cases (78/104, 75.00%), while 26 cases had no tears (26/104, 25.00%). The distribution of tear subtypes was as follows: articular-sided tears in 20 cases (20/104, 19.23%), bursal-sided tears in 13 cases (13/104, 12.50%), intratendinous tears in 20 cases (20/104, 19.23%), and full-thickness tears in 26 cases (26/104, 25.00%).

Comparison of diagnostic value for supraspinatus tendon tears

Diagnostic value of supraspinatus tendon tears for MSKUS, MRI, and USMRA

In diagnosing whether the supraspinatus tendon was torn or not, the diagnostic sensitivity, specificity, positive predictive value, negative predictive value, and accuracy for MSKUS were 83.87%, 67.31%, 88.44%, 58.33%, and 79.71%, respectively (Table 1), with an area under the receiver operating characteristic (ROC) curve (AUC) of 0.76 (see Figure 3A); those for MRI were 92.21%, 80.77%, 93.42%, 77.78%, and 89.32%, respectively, with an AUC of 0.86 (Figure 3B); and those for USMRA were 97.44%, 92.31%, 97.44%, 92.31%, and 96.15%, respectively, with an AUC of 0.95 (Figure 3C). These results represented significant differences between the three groups, with all differences being statistically significant (P<0.05). The ROC curve indicated that MSKUS and MRI had moderate diagnostic value, while USMRA had high diagnostic value.

Figure 3 ROC curves for the presence or absence of supraspinatus tendon tears of MSKUS, MRI, and USMRA. (A) MSKUS; (B) MRI; (C) USMR. ROC, receiver operating characteristic; AUC, area under the receiver operating characteristic curve; MSKUS, musculoskeletal ultrasound; MRI, magnetic resonance imaging; USMRA, ultrasound-guided magnetic resonance arthrography.

Kappa consistency analysis of MSKUS, MRI, USMRA, and shoulder arthroscopy in determining tearing of the supraspinatus tendon

The kappa value between MSKUS and arthroscopy was 0.49 (95% CI: 0.35–0.62; P<0.0001), while that between MRI and arthroscopy was 0.72 (95% CI: 0.57–0.88; P<0.0001), both indicating moderate consistency. USMRA showed good consistency with arthroscopic diagnosis, with a kappa value of 0.89 (95% CI: 0.79–0.99; P<0.0001), as shown in Table 2.

Comparison of diagnostic value for tear subtypes

Articular-sided supraspinatus tendon tears

The diagnostic the diagnostic sensitivity, specificity, positive predictive value, negative predictive value, and accuracy for MSKUS were 57.14%, 67.31%, 54.05%, 70.00%, and 63.22%, respectively; those for MRI were 66.67%, 84.00%, 71.43%, 80.77%, and 77.50%, respectively; and those for USMRA were 85.00%, 88.89%, 85.00%, 88.89%, and 87.23%, respectively; meanwhile, the AUCs were 0.62, 0.75, and 0.87, respectively, indicating moderate diagnostic value, as shown in Figure 4A-4C. The above results suggested significant differences in the diagnostic sensitivity, specificity, positive predictive value, and accuracy for articular-sided supraspinatus tendon tears between MSKUS, MRI, and USMRA (P<0.05). However, there was no significant difference in the negative predictive value (P>0.05), as shown in Table 3. The results of ROC curve suggested that the subtype diagnostic value for articular-sided supraspinatus tendon tears for MSKUS was low, while that of MRI and USMRA was moderate.

Figure 4 ROC curves for the subtypes of articular-sided supraspinatus tendon tears of MSKUS, MRI, and USMRA. (A) MSKUS; (B) MRI; (C) USMR. ROC, receiver operating characteristic; AUC, area under the curve; MSKUS, musculoskeletal ultrasound; MRI, magnetic resonance imaging; USMRA, ultrasound-guided magnetic resonance arthrography.

Bursal-sided supraspinatus tendon tears

The diagnostic sensitivity, specificity, positive predictive value, negative predictive value, and accuracy for MSKUS were 65.63%, 76.09%, 65.63%, 76.09% and 71.79%, respectively; those for MRI were 71.43%, 75.00%, 68.18%, 77.78%, and 73.47%, respectively; and those for USMRA were 91.67%, 92.31%, 84.62%, 96.00%, and 92.11%, respectively. These findings indicated no significant differences (P>0.05), except for accuracy (P=0.02), between the MSKUS, MRI, and USMRA groups for bursal-sided supraspinatus tendon tears (Table 4); moreover, the AUCs for the three groups were 0.71, 0.73, and 0.92, respectively, indicating moderate diagnostic value for MSKUS and MRI and a high diagnostic value for USMRA (Figure 5A-5C). The results indicated that the subtype diagnostic values for bursal-sided supraspinatus tendon tears for MSKUS and MRI were moderate, while that for USMRA was high.

Figure 5 ROC curves for bursal-sided supraspinatus tendon tears of MSKUS, MRI, and USMRA. (A) MSKUS; (B) MRI; (C) USMR. ROC, receiver operating characteristic; AUC, area under the curve; MSKUS, musculoskeletal ultrasound; MRI, magnetic resonance imaging; USMRA, ultrasound-guided magnetic resonance arthrography.

Intratendinous supraspinatus tendon tears

The diagnostic sensitivity, specificity, positive predictive value, negative predictive value, and accuracy for MSKUS were 52.63%, 61.40%, 47.62%, 66.04%, and 57.89%, respectively; those for MRI were 68.42%, 75.00%, 65.00%, 77.78%, and 72.34%, respectively; and those for USMRA were 89.47%, 88.89%, 85.00%, 92.31%, and 89.13%, respectively. These findings indicated significant differences between the MSKUS, MRI, and USMRA groups (P<0.05), as shown in Table 5; moreover, the AUCs were 0.57, 0.72, and 0.89, respectively, indicating moderate diagnostic value, as shown in Figure 6A-6C. The results showed that the subtype diagnostic value for intratendinous supraspinatus tendon tears for MSKUS was low, while that of MRI and USMRA was moderate.

Figure 6 ROC curves for intratendinous supraspinatus tendon tears of MSKUS, MRI, and USMRA. (A) MSKUS; (B) MRI; (C) USMR. ROC, receiver operating characteristic; AUC, area under the curve; MSKUS, musculoskeletal ultrasound; MRI, magnetic resonance imaging; USMRA, ultrasound-guided magnetic resonance arthrography.

Full-thickness supraspinatus tendon tears

The diagnostic sensitivity, specificity, positive predictive value, negative predictive value, and accuracy for MSKUS were 62.00%, 87.50%, 86.11%, 64.81%, and 73.33%, respectively; those for MRI were 81.82%, 91.30%, 90.00%, 84.00%, and 86.67%, respectively; and those for USMRA were 89.29%, 96.00%, 96.15%, 88.89%, and 92.45%, respectively. These findings indicated significant differences in between MSKUS, MRI, and USMRA (P<0.05), except in specificity and positive predictive value (P>0.05), as shown in Table 6. The AUCs for MSKUS, MRI, and USMRA were 0.75, 0.87, and 0.93, respectively, indicating moderate diagnostic value for MSKUS and MRI and high diagnostic value for USMRA (Figure 7A-7C). The results demonstrated that the subtype diagnostic value for full-thickness supraspinatus tendon tears of MSKUS and MRI was moderate, while that of USMRA was high.

Figure 7 ROC curves for full-thickness supraspinatus tendon tears of MSKUS, MRI, and USMRA. (A) MSKUS; (B) MRI; (C) USMR. ROC, receiver operating characteristic; AUC, area under the curve; MSKUS, musculoskeletal ultrasound; MRI, magnetic resonance imaging; USMRA, ultrasound-guided magnetic resonance arthrography.

Success rate and adverse reactions of ultrasound-guided shoulder joint puncture

In 104 cases, the operation of USMRA was successful in 103 cases, representing a success rate of 99.04% (103/104). One patient experienced vagal reflex including excessive sweating and dizziness during the injection of the contrast agent. The puncture procedure was stopped, and the patient was placed in a supine position and rested for 10 minutes before the puncture was resumed. One patient had bruising at the injection site. The incidence of adverse reactions was 1.92% (2/104), and there were no other significant adverse reactions.

Discussion

Imaging techniques such as MSKUS and MRI are important methods for diagnosing and differentiating supraspinatus tendon injuries. MSKUS is widely used in clinical practice due to its cost-effectiveness and convenience. However, the diagnostic performance of MSKUS varies widely in the literature, with sensitivities ranging from 73.1% to 94.3%, specificities ranging from 62.5% to 86.4%, and accuracies ranging from 70.6% to 92.9%. This may be attributed to the scanning technique, skill, and clinical experience of the doctors (15,16). In our study, MSKUS had a sensitivity of 83.87%, a specificity of 67.31%, a positive predictive value of 88.44%, a negative predictive value of 58.33%, and an accuracy of 79.71% in diagnosing supraspinatus tendon tears, which is consistent with the results of Liu et al.’s study (17). The AUC for MSKUS was 0.76, suggesting that MSKUS has moderate diagnostic value in identifying the presence of supraspinatus tendon tears.

As an important imaging technique, MRI can complement MSKUS by detecting tears and fat infiltration that may be missed by MSKUS, providing important evidence for clinical treatment. Zhang et al. reported that MRI had a sensitivity of 93.3%, a specificity of 88.9%, and an accuracy of 92.8% in diagnosing supraspinatus tendon tears, with no significant difference compared to MSKUS (18). We further found that MRI had a sensitivity of 90.41%, a specificity of 80.77%, and an accuracy of 87.88% in diagnosing supraspinatus tendon tears, with an AUC of 0.86, indicating moderate diagnostic value.

Traditional MRA, which involves blind puncture, requires high technical proficiency and may lead to complications such as misplaced injections or injection into surrounding tissues, influencing the quality of MRI images and potentially causing severe complications such as pneumothorax and vascular or neural injury (6-8). However, if the gadolinium contrast agent can be successfully injected into the glenohumeral joint cavity or subacromial bursa under ultrasound guidance, it can be infused into the injured area along the supraspinatus tendon tear, facilitated the clear visualization of the tear and improving the diagnostic efficacy of MRI. Our study demonstrated that ultrasound-guided injection of gadolinium contrast agents had a high success rate (99.04%) and a low incidence of adverse reactions. USMRA yielded a sensitivity of 97.44%, a specificity of 92.31%, a positive predictive value of 97.44%, a negative predictive value of 92.31%, and an accuracy of 96.15% in diagnosing supraspinatus tendon tears, with an AUC of 0.95, indicating high diagnostic value.

The subtypes of supraspinatus tendon tears play a crucial role in determining clinical treatment plans. Full-thickness tears may require surgical treatment, while other partial tears require different surgical repair methods (11,19). Particularly for patients who do not respond well to conservative treatment and require ultrasound-guided intervention, accurate subtyping of supraspinatus tendon tears is vital for selecting the appropriate puncture route (12,20,21).

In our study, MSKUS had moderate clinical value in diagnosing the presence of supraspinatus tendon tears, accompanied by moderate consistency with arthroscopic examination. However, MSKUS had limited value in finely classifying supraspinatus tendon tears, particularly for articular-sided tears (AUC =0.62) and intratendinous tears (AUC =0.57), suggesting low clinical diagnostic value. This could be due to the deep location of articular-sided and intratendinous tears, which makes MSKUS susceptible to muscle tendon anisotropy artifacts that can have an effect on the diagnosis. Bursal-sided tears are located in a shallow position and have high resolution, resulting in clear and distinguishable images. In contrast, for Bursal-sided and full-thickness tears, the AUCs were 0.71 and 0.75, respectively, indicating moderate diagnostic value for MSKUS.

MSKUS examination of the supraspinatus tendon requires a high technical ability from the operator and specialized positioning from the patient, with the modified cross position commonly being used. In this position, the patient’s hands are inside the pockets, the shoulder joint is projected forward, and the elbow is tilted back, exposing the supraspinatus tendon in the subacromial space for ultrasound examination. The 1-cm area attaching the supraspinatus tendon to the greater tuberosity should be carefully scanned, and the probe should be properly tilted in order to keep the sound speed perpendicular to the tendon and avoid anisotropy artifacts. Obtaining the optimal cross-sectional image of supraspinatus tendon is key to improving the diagnostic performance for subtyping with MSKUS (22,23).

There is significant variability in the measures for MRI-based subtyping of supraspinatus tendon tears in the literature, with sensitivities ranging from 86.3% to 94.4%, specificities ranging from 58.3% to 67.31%, and accuracies ranging from 70.87% to 78.5% (24,25). In our study, the sensitivities ranged from 66.67% to 81.82%, the specificities ranged from 75.00% to 91.30%, and the accuracies ranged from 72.34% to 86.67% for identifying the subtypes of articular-sided, bursal-sided, intratendinous, and full-thickness tears on MRI, representing moderate diagnostic value. The acute phase of supraspinatus tendon tear is often accompanied by joint effusion, and MRI examination in such cases can improve the detection rate and facilitate subtyping. This may contribute to the significant variation in MRI-based subtyping of supraspinatus tendon tears found in the literature.

The key to MRA technology is a precise injection of gadolinium contrast agent (26). Under normal physiological conditions, the glenohumeral joint cavity is not connected with the subacromial bursa, but the two are connected in cases of full-thickness supraspinatus tendon tears. Therefore, it is particularly important to evaluate the presence and amount of fluid in the subacromial bursa before injection of contrast agent in MRA. If there is no or minimal fluid in the subacromial bursa before contrast agent injection but fluid appears or increases in the subacromial bursa after guided injection into the glenohumeral joint under ultrasound, this suggests a full-thickness tear. In contrast, traditional MRA does not fully evaluate the subacromial bursa before contrast agent injection, so it is difficult to differentiate full-thickness tears when both the subacromial bursa and glenohumeral joint show fluid, leading to a possible misdiagnosis of full-thickness supraspinatus tendon tears. In the USMRA group, our study showed a puncture success rate of 99.04% (103/104) and an adverse reaction rate of 1.92% (2/104) for ultrasound-guided injection into the glenohumeral joint and subacromial bursa for USMRA. This proves that ultrasound guidance can effectively elevate the success rate of puncture and reduce complications such as nerve, vascular, and pneumothorax injuries. Furthermore, the diagnostic sensitivity, specificity, positive predictive value, negative predictive value, and accuracy of full-thickness tears were 89.29%, 96.00%, 96.15%, 88.89%, and 92.45%, respectively, which were significantly higher than those in the MSKUS and MRI groups (P<0.05). Moreover, the AUC of MSKUS and MRI were 0.75 and 0.87 respectively, while the AUC of USMRA was 0.93, which suggested that the USMRA had higher diagnostic value for full-thickness tears than did MSKUS and MRI.

The novelty of this study is the particular leveraging of ultrasound visualization technology, which not only improved the success rate of puncture but also greatly reduced the complications. Moreover, compared with traditional MRA single-chamber arthrography of the glenohumeral joint, USMRA can facilitate dual-chamber arthrography. Ultrasound-guided puncture of the shoulder joint not only accurately punctures the glenohumeral joint cavity but also accurately punctures the subacromial bursa, which can effectively improve the diagnostic performance of tear subtypes.

The limitations of this study were as follows: (I) because patients were enrolled from only a single hospital, the sample size of this study was small; (II) there were individual differences in sensitivity to pain among patients, which could have affected patients’ choice of examination methods, (III) supraspinatus tendon tears are often accompanied by limited mobility, and poor positioning during MSKUS examination might have caused interference from tendon anisotropy artifacts, resulting in biased MSKUS diagnostic results. We will further expand the sample size for in-depth research in the future.

Conclusions

In the diagnosis of supraspinatus tendon tears, USMRA demonstrated a higher diagnostic value than did MSKUS and MRI, with an AUC of 0.95, in addition to better consistency with arthroscopic examination. Moreover, in subtype diagnosis of supraspinatus tendon tears, USMRA is valuable in the diagnosis of bursal-sided and full-thickness tears.

Acknowledgments

Funding: This study was funded by Medical and Health Science and Technology Project of Zhejiang Province (No. 2022RC263) and the Huzhou University “Four New” Education and Teaching Reform Research Project (No. JG202262).

Footnote

Reporting Checklist: The authors have completed the STARD reporting checklist. Available at https://qims.amegroups.com/article/view/10.21037/qims-24-765/rc

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://qims.amegroups.com/article/view/10.21037/qims-24-765/coif). The authors have no conflict of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. This study was conducted in accordance with the Declaration of Helsinki (as revised in 2013) and was approved by the Medical Ethics Committee of Huzhou Central Hospital (No. 202206043-01). All patients provided their written informed consent.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Muench LN, Berthold DP, Otto A, Dyrna F, Bell R, Obopilwe E, Cote MP, Imhoff AB, Mazzocca AD, Beitzel K. Increased Glenohumeral Joint Loads Due to a Supraspinatus Tear Can Be Reversed With Rotator Cuff Repair: A Biomechanical Investigation. Arthroscopy 2022;38:1422-32. [Crossref] [PubMed]
2. Stengaard K, Hejbøl EK, Jensen PT, Degn M, Ta TML, Stensballe A, Andersen DC, Schrøder HD, Lambertsen KL, Frich LH. Early-stage inflammation changes in supraspinatus muscle after rotator cuff tear. J Shoulder Elbow Surg 2022;31:1344-56. [Crossref] [PubMed]
3. Ellman H. Diagnosis and treatment of incomplete rotator cuff tears. Clin Orthop Relat Res 1990;64-74. [Crossref] [PubMed]
4. Kim JH, Park JW, Heo SY, Noh YM. Magnetic resonance imaging analysis of rotator cuff tear after shoulder dislocation in a patient older than 40 years. Clin Shoulder Elb 2020;23:144-51. [Crossref] [PubMed]
5. Zhang X, Gu X, Zhao L. Comparative Analysis of Real-Time Dynamic Ultrasound and Magnetic Resonance Imaging in the Diagnosis of Rotator Cuff Tear Injury. Evid Based Complement Alternat Med 2021;2021:2107693. [Crossref] [PubMed]
6. Omer T, Perez M, Berona K, Lam CN, Sajed D, Brandon C, Falkenstein J, Kang T, Mailhot T. Accuracy of Landmark-guided Glenohumeral Joint Injections as Assessed by Ultrasound in Anterior Shoulder Dislocations. West J Emerg Med 2021;22:1335-40. [Crossref] [PubMed]
7. Raeissadat SA, Rayegani SM, Langroudi TF, Khoiniha M. Comparing the accuracy and efficacy of ultrasound-guided versus blind injections of steroid in the glenohumeral joint in patients with shoulder adhesive capsulitis. Clin Rheumatol 2017;36:933-40. [Crossref] [PubMed]
8. Chung CB, Dwek JR, Feng S, Resnick D. MR arthrography of the glenohumeral joint: a tailored approach. AJR Am J Roentgenol 2001;177:217-9. [Crossref] [PubMed]
9. Chae SH, Won JY, Yoo JC. Clinical outcome of ultrasound-guided atelocollagen injection for patients with partial rotator cuff tear in an outpatient clinic: a preliminary study. Clin Shoulder Elb 2020;23:80-5. [Crossref] [PubMed]
10. Tang YQ, Zeng C, Su XT, Li SS, Yi WH, Xu JJ, Wu GQ, Chen YJ, Li MW, Liu HM. The Value of Percutaneous Shoulder Puncture with Contrast-enhanced Ultrasound in Differentiation of Rotator Cuff Tear Subtypes: A Preliminary Prospective Study. Ultrasound Med Biol 2019;45:660-71. [Crossref] [PubMed]
11. Donohue NK, Nickel BT, Grindel SI. High-Grade Articular, Bursal, and Intratendinous Partial-Thickness Rotator Cuff Tears: A Retrospective Study Comparing Functional Outcomes After Completion and Repair. Am J Orthop (Belle Mead NJ) 2016;45:E254-60. [PubMed]
12. Huang SH, Hsu PC, Wang KA, Chou CL, Wang JC. Comparison of single platelet-rich plasma injection with hyaluronic acid injection for partial-thickness rotator cuff tears. J Chin Med Assoc 2022;85:723-9. [Crossref] [PubMed]
13. Shen J. Evaluation of specific signs and indirect magnetic resonance arthrography in the diagnosis of supraspinatus injury and the research of arthroscopic treatment. Suzhou: Soochow University; 2017.
14. Jordan RW, Naeem R, Srinivas K, Shyamalan G. A comparison of magnetic resonance arthrography and arthroscopic findings in the assessment of anterior shoulder dislocations. Skeletal Radiol 2015;44:653-7. [Crossref] [PubMed]
15. Lei HL. The Clinical Application Value of High Frequency Ultrasound and Shear Wave Elastography in Rotator Cuff. Nanchong: North Sichuan Medical College; 2021.
16. Xu TL. The value of high frequency ultrasound and MRI in the diagnosis of partial and full-thickness rotator cuff tears. Chengde: Chengde Medical University; 2020.
17. Liu F, Dong J, Shen WJ, Kang Q, Zhou D, Xiong F. Detecting Rotator Cuff Tears: A Network Meta-analysis of 144 Diagnostic Studies. Orthop J Sports Med 2020;8:2325967119900356. [Crossref] [PubMed]
18. Zhang LL, Chen YY, Wang HJ, Chen L. Values of High-Frequency Ultrasound and MRI in Diagnosis of Supraspinatus Tendon Tears. Practical Clinical Medicine 2019;20:54-7.
19. Zhang ZY, Wang HB, Lou XY, Miao BJ, Zhou S, Wang C, Meng J. The clinical application of MR shoulder arthrography in the diagnosis of complete tear of rotator cuff. Radiologic Practice 2016;31:531-4.
20. Berná-Mestre JD, Fernández C, Carbonell G, García A, García-Vidal JA, Mirapeix FMI, Berná-Serna JD. Influence of Acromial Morphologic Characteristics and Acromioclavicular Arthrosis on the Effect of Platelet-Rich Plasma on Partial Tears of the Supraspinatus Tendon. AJR Am J Roentgenol 2020;215:954-62. [Crossref] [PubMed]
21. Thepsoparn M, Thanphraisan P, Tanpowpong T, Itthipanichpong T. Comparison of a Platelet-Rich Plasma Injection and a Conventional Steroid Injection for Pain Relief and Functional Improvement of Partial Supraspinatus Tears. Orthop J Sports Med 2021;9:23259671211024937. [Crossref] [PubMed]
22. Jacobson JA. Shoulder US: anatomy, technique, and scanning pitfalls. Radiology 2011;260:6-16. [Crossref] [PubMed]
23. Kim H, Park HJ, Lee SY, Kim JN, Moon J, Kim MS, Kim E. Ultrasound evaluation of postsurgical shoulder after rotator cuff repair: comparison of clinical results. Acta Radiol 2021;62:1025-34. [Crossref] [PubMed]
24. Wang HB, Zhang ZY, Lou XY, Miao BJ, Song YF, Sun SF, Wang C, Meng J. The clinical application of MR shoulder arthrography in rotator cuff injury. Journal of Medical Imaging 2016;26:1683-7.
25. Yashengjiang Yusufu. Comparative analysis of supraspinatus tendon injury with magnetic resonance imaging and shoulder arthroscopy. Urumqi: Xinjiang Medical University; 2021.
26. Shang PY, Yu LC, Zhang SL, Cheng XY, Shi JX, Shi WJ. Arthroscopic Suture-bridge Technique for Minimally Invasive Treatment of Intratendinous Partial- thickness Rotator Cuff Tears. Chinese Journal of Sports Medicine 2022;41:602-11.