Clinical Journal of Sport Medicine

Clinical Journal of Sport Medicine

Posted on by admin


INTRODUCTION

Adhesive capsulitis, commonly referred to as “frozen shoulder,” is a disabling and painful condition that affects approximately 2% to 5% of the general population,1 although estimates range up to 10%.2 The incidence is higher in female patients3 and individuals between the age of 40 and 65 years.4 While the etiology of primary adhesive capsulitis is largely unknown, numerous conditions have been associated with its development including diabetes mellitus,5 thyroid disorders,6,7 hyperlipidemia,8 autoimmune disorders,9 smoking,10 obesity,11 and low physical activity level.12 Shoulder trauma, surgery, and prolonged immobilization are each risk factors of secondary adhesive capsulitis.13

Adhesive capsulitis classically presents with shoulder pain and gradual loss of both active and passive range of motion (ROM). The natural history can be characterized by 3 distinct clinical stages.14 First, the “freezing phase” occurs with onset of shoulder pain and preserved ROM. Second, the “frozen phase” follows where pain may improve, but ROM is restricted, most notably shoulder external rotation. Finally, mobility gradually returns in the “thawing phase” with typical resolution over 3 phases ranging from 1 to 3 years. While it is often regarded as a self-limiting condition, incomplete restoration of ROM is common.15

The pathophysiology is thought to involve both inflammatory and fibrotic processes in the joint capsule, resulting in thickening of the adjacent synovial membrane, contracture of the capsule, and reduced capsular volume.14,16,17 Arthroscopic studies have demonstrated that the predominant site of pathology in adhesive capsulitis is at the anterior and axillary joint capsule, coracohumeral ligament, and the rotator interval.18–20 The rotator interval is located between the supraspinatus and subscapularis tendons. It is a triangular space that contains the anterosuperior portion of the glenohumeral capsule, superior glenohumeral ligament, coracohumeral ligament, and long head of the biceps tendon.21

Adhesive capsulitis is associated with significant pain and functional impairment. Initially, most patients with adhesive capsulitis are managed conservatively with steroid injections, medications, and physical therapy, although an evidence-based treatment algorithm has yet to be defined.22 A recent systematic review concluded that intra-articular corticosteroid injection may be superior to other nonsurgical treatments and further improvement results when combined with a home exercise program after injection.23 Moreover, steroid injection combined with hydrodilatation, which provides an immediate expansion of the capsule, may potentially expedite recovery.24

Glenohumeral joint steroid injections are commonly described using a posterior approach.25 However, some have suggested that an anterior injection approach targeting the rotator interval may allow for a more targeted and specific treatment based on the anatomy and observed pathology in adhesive capsulitis.26 Several randomized controlled trials (RCTs) comparing outcomes between the anterior and posterior approaches have yielded conflicting results.27–32 Therefore, the purpose of this systematic review and meta-analysis was to compare outcomes between the anterior and posterior glenohumeral injection approaches. The primary outcomes were pain visual analog scale (VAS) and shoulder ROM at 12 weeks, with secondary outcomes assessing patient-reported functional outcomes, accuracy, and adverse events. Given the current known pathophysiology of this condition, we hypothesized that the anterior approach would be associated with more pain reduction and increased ROM compared with the posterior approach.

METHODS

Systematic Review Registration

The protocol for this systematic review was registered at International Platform of Registered Systematic Review and Meta-Analysis Protocols (INPLASY202310080).

Search Strategy and Selection Criteria

We searched MEDLINE, Embase, Web of Science, and Cochrane Center Register of Controlled Trials for RCTs and prospective comparative studies that compared anterior and posterior glenohumeral joint approaches for steroid injection to treat adhesive capsulitis. The literature search was conducted per the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines and included all studies published from database inception through February 2023.33 A librarian affiliated with the authors’ institution was involved in building search terms (see Appendix A, Supplemental Digital Content, https://links.lww.com/JSM/A408). Reference lists of relevant articles were also used for a manual search. Randomized controlled trials or prospective comparative studies that compared outcomes between anterior and posterior approaches with steroids for adhesive capsulitis were eligible for inclusion.

Briefly, the anterior approach was defined as an injection into the superior–anterior glenohumeral joint capsule. In the majority of studies included,27–29,31,32 needle placement was confirmed under ultrasound guidance, targeting just under the coracohumeral ligament either at the rotator interval or near the coracoid process (Figure 1A). Two articles30,34 used landmark-guided anterior glenohumeral joint injections, where the needle was inserted at the soft spot between the anterior acromion and coracoid, and advanced until contact with the humeral head was made. Posterior approach was defined as an injection into the glenohumeral space through the posterior deltoid (Figure 1B). In the ultrasound guided method, the space between the humeral head and glenoid labrum was visualized, and the injection was made into this space.27–29,31,32 In the landmark-guided posterior approach,30,34 the needle was inserted at the soft point, 2 to 3 cm inferomedial to the posterolateral acromion and advanced toward the coracoid process until contact with the humeral head was made. For the procedures, 21 to 23 gauge and 6-cm needles were used in the included studies.

Figure 1.:

Ultrasound-guided injection through the anterior rotator interval approach (A), posterior approach (B), postinjection extra-articular biceps tendon sheath flow (C), postinjection glenohumeral joint arthrogram (D). White arrows are pointing toward the injection target.

Articles written in English or in languages that could be translated to English were included. Studies in which anterior glenohumeral joint injections were combined with posterior/inferior glenohumeral joint injections, or extra-articular injections (subacromial bursa, biceps tendon, etc.) were excluded. Based on these criteria, 2 trials were excluded from our analyses because one study utilized 2 injections and targeted the coracohumeral ligament and inferior glenohumeral capsule,35 and the other study compared combined anterior and posterior injections and to the posterior approach alone.26 Furthermore, retrospective studies, case series, and case reports were excluded.

Data Extraction and Quality Assessment

Two authors independently evaluated each study from the initial search. Data extraction was independently conducted by 2 authors. Variables extracted from each article included country where the study was performed, study design, study inclusion/exclusion criteria, injection details (eg, ultrasound-guidance, hydrodilatation, dose, number of injections), rehabilitation protocols, follow-up periods, and adverse events. Patient characteristics including age, sex, laterality, arm dominance, and duration of symptoms were collected. The mean and SD of outcome measures after anterior and posterior shoulder injection were extracted for statistical analysis. When values were unavailable from the published manuscript, study corresponding authors were contacted for additional data. Two authors assessed the risk of bias with the revised Cochrane risk-of-bias tool for randomized trials for RCTs36 and the Newcastle–Ottawa Scale (NOS) for non-RCT.37 Discrepancies between 2 authors were resolved by discussion with a third author.

Outcome Measures

The primary outcomes of this meta-analysis were pain VAS and passive shoulder ROM (forward flexion, abduction, external rotation, and internal rotation) at 12 weeks. Secondary outcomes were pain VAS at 8 weeks, American Shoulder and Elbow scores (ASES) at 12 weeks, Constant–Murley shoulder score (Constant score), Shoulder Pain and Disability Index (SPADI) at 12 weeks, accuracy, and adverse events.

Statistical Analysis

We performed random-effects pairwise meta-analyses to account for clinical heterogeneity across the studies including patient and injection characteristics. Statistical heterogeneity was assessed with the Q and I2 statistics.38 The standardized mean difference (SMD) between anterior and posterior approaches was used as a measure of effect size for patient-reported outcomes including pain VAS, ASES, Constant score, and SPADI for better interpretability.39 Pain at rest was used for our meta-analysis.32 An SMD of 0.2 represented a small effect size, 0.5 represented a medium effect size, and 0.8 or larger represented a large effect size. For ROMs, the weighted mean difference (WMD) was used as a measure of effect size because the degrees of ROMs are directly interpretable. For accuracy, a meta-analysis of proportion was conducted. Subgroup analyses were performed with studies that used corticosteroids without hydrodilatation and studies that used ultrasound-guidance for injections. Publication bias was not assessed because the included number of studies was less than 10. STATA Version 16 (StataCorp, LLC, College Station, TX) was used for all analyses.

RESULTS

Eligible Studies and Characteristics of Included Studies

The initial search of published articles yielded 1066 studies. Three hundred sixty duplicate articles were removed. Of 706 articles reviewed, 676 articles met exclusion criteria based on review of titles and abstracts. Thirty studies were reviewed in full, and 23 studies were excluded based on ineligible intervention, study design, comparator, or patient population. The resulting 7 articles identified in the search were included in the meta-analysis. The review process is outlined in the PRISMA flow diagram (Figure 2).

F2
Figure 2.:

PRISMA (preferred reporting items for systematic meta-analyses) diagram showing selection of articles for this meta-analysis.

Of the studies included, 6 were RCTs, and one study was a sequential prospective cohort study. Together, the studies comprised 468 patients, of which 241 patients received anterior injections, and 227 patients received posterior glenohumeral joint injections. The sample sizes of specific studies ranged from 23 to 54 patients in either the anterior or posterior injection group. The majority of patients were in their 50s, and most patients were female (n = 60%). Inclusion criteria for all studies required that patients have a primary diagnosis of adhesive capsulitis without secondary causes of shoulder pain based on clinical findings, imaging, or both. In addition, 5 studies had criteria for duration of shoulder pain ranging from more than one month to more than 9 months. Patients received ultrasound guided injections in 5 studies.27–29,31,32 In all 7 studies, patients received one injection containing between 3 and 20 mL of injectate comprised corticosteroid (40 mg triamcinolone in 5 studies or 40 mg methylprednisolone in 2 studies), lidocaine, normal saline and/or contrast. Three studies utilized hydrodilatation along with steroid injections when comparing the 2 approaches.28,29,32 Hydrodilatation was performed for both the anterior and posterior approaches in these 3 studies.

Following the injection, 5 studies provided patients with home exercise regimens, one study provided patients with physical therapy–guided exercises, and one study did not report a rehabilitation program. Patients were followed for 12 weeks after injection at various intervals.

Further characteristics of individual studies are summarized in Table 1.


TABLE 1. -
Summary of Individual Study Characteristics












Study Year Country Study Design Inclusion Criteria Anterior Injection Posterior Injection Ultrasound Hydrodilatation Injection Rehabilitation Outcome Measures Follow-Up Adverse Events
Patients (n, Mean Age, yr ± SD) Female (%) BMI (Mean ± SD) Patients (n, Mean Age, yr ± SD) Female (%) BMI (Mean ± SD)
Cho et al27 2022 South Korea RCT 1) Shoulder pain with limitation of passive motion of >30° in 2 or more movement planes at the time of presentation
2) Plain radiography and magnetic resonance imaging to rule out secondary causes of painful stiffness		 43, 54.1 ± 8.9 41.9 23.6 ± 3.0 45, 55.4 ± 9.9 55.6 23.7 ± 2.6 Yes No 1 mL (40 mg/mL) triamcinolone acetonide, 3 mL 1% lidocaine, 3 mL of water-soluble unionized contrast, 3 mL of normal saline (total volume: 10 mL) Home-based exercise program with pendulum exercises, wall-climbing exercises, and gentle ROM exercises using a bar (3 times/day, 15 minutes/round). Patients were asked to refrain from provoking postmobilization soreness with self-feedback Primary: Pain VAS score, ASES score, SSV, passive ROMs 3, 6, and 12 wk No patients reported serious adverse effects
Deng et al30 2023 China RCT 1) Age older than 18 years with unilateral shoulder pain and limited movement diagnosed with primary frozen shoulder
2) Duration of pain >9 months and VAS score >3
3) Limitation of active and passive movement of the affected shoulder joint by > 30° relative to the contralateral shoulder joint in at least
2 movement planes
4) Routine shoulder x-ray and/or MRI to exclude other pathologies		 30, 54.5 ± 5.7 76.7 23.8 ± 2.5 30, 55.3 ± 3.9 66.7 24.5 ± 2.5 No No 1 mL (40 mg/mL) triamcinolone acetonide, 4 mL 2% lidocaine (total volume: 5 mL) Self-exercise program composed of flexion, abduction, external rotation and internal rotation under mild active training, and each exercise was performed 15-20 times during a period (3 times/d, 15 min/period). Patients were excluded if the exercise was not strictly performed during the 12-wk follow-up period Primary: Pain VAS, constant–Murley shoulder score
Secondary: Passive ROM and complication		 4, 8 and 12 wk One patient in the anterior injection group experienced temporary symptoms associated with needle syncope
Elnady et al28 2020 Egypt RCT 1) Age 35 to 60 yrs
2) Pain and stiffness in only one shoulder, for 1 to 6 mo
3) Restriction of passive motion, as measured with goniometer		 32, 45.4 ± 4.9 75.0 — 32, 47.6 ± 13.5 71.9 — Yes Yes 1 mL (40 mg/mL) methyl-prednisolone acetate,
1 mL 2% lidocaine, 15 mL normal saline (total volume: 17 mL)		 Physiotherapist-guided stretching and strengthening exercise program, every other day for 12-wk follow-up period Primary: Pain VAS, SPADI 12 wk Seven patients (3 with posterior approach and 4 with anterior approach) experienced transient local pain and facial flushing following injection. Three patients (2 with posterior approach and 1 with anterior approach) required pain relief with NSAIDs for 72 h following injection
Kim et al29 2017 South Korea RCT 1) Shoulder pain and limitations of active and passive motion in at least 2 directions on the initial checkup
2) MRI or ultrasonography with no secondary cause for adhesive capsulitis		 23, 55.5 ± 11.1 56.5 — 23, 52.8 ± 7.6 69.6 — Yes Yes 1 mL (40 mg/mL) triamcinolone, 4 mL lidocaine, 4 mL normal saline, 3 mL omnipaque (total volume: 13 mL) Stick exercises started at 3 wks, and TheraBand exercises started at 7 wks after injection Primary: Pain VAS, ROM, SAT, ASES, constant–Murley shoulder score
Secondary: Consumed time		 3, 7, and 13 wk Not reported
Rijs et al34 2021 Netherlands S-PCT 1) Clinical symptoms 2) Shoulder pain with impaired passive glenohumeral ROM of ≥30° in 2 or more planes
3) Radiographs and ultrasound or MRI to confirm diagnosis and rule out other causes of shoulder pain		 54, 54.0 ± 4.0 63.0 23.0 ± 3.0 41, 52.0 ± 5.0 58.5 24.0 ± 3.0 No No 1 mL (40 mg/mL) methylprednisolone, 2 mL (10 mg/mL) lidocaine, 3 mL (320 mg L/mL) loxaglic acid (total volume: 6 mL) Not reported Primary: Accuracy of the injection (contrast fluid localized in the glenohumeral joint without leakage of contrast fluid in the soft tissue or subacromial space) N/A Not reported
Sun et al25 2017 China RCT 1) Age 18 to 70 yrs
2) Primary frozen shoulder
3) Normal radiograph finding of the affected shoulder
4) Loss of passive motion of the glenohumeral joint >25% or 30° in at least 2 planes when compared with the contralateral side or normal value
5) Pain VAS >7
6) Duration of symptoms 		27, 52.6 ± 4.4 59.3 — 24, 55.1 ± 3.4 58.3 — Yes No 1 mL (40 mg/mL) triamcinolone, 2 mL 2% lidocaine (total volume: 3 mL) Home regimen of daily shoulder exercises Primary: Pain VAS
Secondary: Constant score, (DASH core, passive ROM)		 4, 8, and 12 wk One patient in each group experienced temporary facial flushing following injection
Wang et al8 2021 Taiwan RCT 1) Age 35 to 65 yrs
2) Duration of symptoms >1 mo
3) Limitation in the passive ROM >30° when compared with the contralateral side in at least 2 of these movements: forward flexion, abduction, or external rotation		 32, 52.4 ± 6.4 62.5 22.9 ± 3.8 32, 54.0 ± 7.0 59.4 24.1 ± 3.1 Yes Yes 4 mL (10 mg/mL) triamcinolone acetonide, 4 mL 2% lidocaine hydrochloride, 12 mL normal saline (total volume: 20 mL) Home exercise program consisted of Codman exercise, wall-climbing exercise, shoulder external and internal rotation exercise using the bar, and towel stretching exercise behind the back. All patients started the exercises at the first day after injections (at least 2 times/d) Patients were
also allowed to receive physical therapy with hot packs, ultrasound diathermy, and inferential current therapy at specific outpatient clinics		 Primary: SPADI score
Secondary: Pain VAS, ROM of the shoulder		 6 and 12 wk One patient in each group endorsed severe pain (VAS > 4) following injection that spontaneously resolved without additional intervention

DASH, disability of arm, hand, and shoulder; PC, posterior capsule approach; RI, rotator interval approach; SAT, patient satisfaction; S-PCT, sequential prospective comparative study; SSV, subjective shoulder value.

Study Quality and Risk of Bias Assessment

Figure 3 displays overall risk of bias assessment of included RCTs. All RCTs used randomized allocation sequence of patients with no baseline difference between groups. Patients in all studies were aware of their assignment as patient positioning during the procedure differs according to the injection approaches. However, there were no deviations from the interventions because patients in the included studies underwent rehabilitation as planned following their procedures. Outcomes were available for nearly all randomized patients throughout the studies except for those who dropped out. In the end, among the 6 included RCTs, 4 studies were judged to have a low risk of bias.27–29,32 One study30 was judged as having a high risk for bias because ASES, which was listed as a primary outcome, was not reported. A separate study did not provide clinical trial registration information, and therefore, we could not assess risk of selection bias.31 The only prospective comparative study was judged to have a low risk of bias.34

F3
Figure 3.:

Risk of bias assessment for randomized controlled trials included in this meta-analysis.

Pain Visual Analog Scale

In all 6 RCTs27–32 included in our meta-analysis reported significant improvement in pain at 12 weeks following interventions with both anterior and posterior approaches. A meta-analysis of these 6 RCTs demonstrated that there was no significant difference in pain VAS at 12 weeks between 2 approaches (Figure 4A, SMD, −0.86; 95% CI, −1.76 to 0.04; I2, 93.8%).

Meta-analyses of randomized controlled trials comparing the anterior and posterior approaches at 12 weeks. A, Pain visual analog scale, (B) flexion, (C) abduction, (D) external rotation, (E) internal rotation. CI, confidence interval; SMD, standardized mean difference; WMD, weighted mean difference.

f4-1
f4-2

Range of Motion

A meta-analysis of 6 RCTs27–32 demonstrated a greater improvement of external rotation with anterior approach (Figure 4B, WMD, 8.08; 95% CI, 0.79-15.38; I2, 88.5%). A meta-analysis of 5 RCTs27,28,30–32 showed a significant difference between the 2 approaches in shoulder abduction at 12 weeks favoring the anterior approach (Figure 4C, WMD, 6.76; 95% CI, 3.05-10.48; I2, 0.0%). However, there were no significant differences in forward flexion (Figure 4D, WMD, 5.86; 95% CI, −0.12 to 11.84; I2 72.5%)27–32 and internal rotation (Figure 4E, SMD, −0.11; 95% CI, −0.35 to 0.13; I2, 14.6%).27–30,32

Secondary Outcomes

A meta-analysis of 2 RCTs30,31 that had data available for pain VAS at 8 weeks showed that there was no significant difference between 2 approaches (see Appendix Figure A, Supplemental Digital Content, https://links.lww.com/JSM/A409, SMD, −2.32; 95% CI, −4.68 to 0.03; I2, 95.1%). A meta-analysis of 2 RCTs27,29 demonstrated that there was no significant difference in ASES at 12 weeks between the 2 approaches (see Appendix Figure B, Supplemental Digital Content, https://links.lww.com/JSM/A410, SMD, −0.26; 95% CI, −0.81 to 0.28; I2, 57.0%) A meta-analysis of 3 RCTs29–31 showed that there was no significant difference in Constant scores at 12 weeks between the 2 approaches (see Appendix Figure C, Supplemental Digital Content, https://links.lww.com/JSM/A411, SMD, 0.98; 95% CI, −0.76 to 2.71; I2, 95.7%). A meta-analysis of 2 RCTs demonstrated that there was no significant difference in SPADI at 12 weeks between the 2 approaches (see Appendix Figure D, Supplemental Digital Content, https://links.lww.com/JSM/A412, SMD, −0.96; 95% CI, −2.77 to 0.84; I2, 95.3%).

Subgroup Analyses

A meta-analysis of 2 RCTs28,32 evaluating steroid injection with hydrodilatation demonstrated a significant difference in pain VAS at 12 weeks favoring the anterior rotator interval approach (see Appendix Figure E, Supplemental Digital Content, https://links.lww.com/JSM/A413, SMD, −0.52; 95% CI, −0.98 to −0.07; I2, 38.1%). A meta-analysis of 4 RCTs27,29–31 investigating a single injection of steroid without hydrodilatation showed that there was no significant difference between the 2 approaches in pain VAS at 12 weeks (see Appendix Figure E, Supplemental Digital Content, https://links.lww.com/JSM/A413, SMD, −1.08; 95% CI, −2.57 to 0.41, I2, 96.2%). Furthermore, a meta-analysis of 5 studies that utilized ultrasound-guidance for injection demonstrated that there was no significant difference in pain VAS between the 2 approaches at 12 weeks (SMD, −0.88; 95% CI, −1.97 to 0.22; I2, 94.8%).

Accuracy

Three studies27,29,34 reported the accuracy of injections with an overall pooled estimate of 87% (95% CI, 76%-98%; I2, 68.8%) for the anterior approach and 87% (95% CI, 78%-97%; I2, 35.6%) for the posterior glenohumeral joint approach.

Adverse Events

The procedures were well tolerated in both approaches without major complications throughout the studies (Table 1). Minor adverse events included transient local pain, perspiration, and facial flushing.

DISCUSSION

This meta-analysis compared outcomes between anterior and posterior glenohumeral joint steroid injections for the treatment of adhesive capsulitis. Based on the available RCTs, the anterior approach resulted in greater improvements in shoulder external rotation and abduction at 12 weeks. While there was no significant difference between the 2 approaches in patient-reported pain and functional outcomes at 12 weeks, subgroup analyses identified that the anterior approach may offer further benefit when steroid injection is combined with hydrodilatation. Both injection approaches had similar accuracy and were tolerated well without major complications.

Patient-Reported Pain and Function

The onset of adhesive capsulitis is associated with an inflammatory hypervascular synovitis, which gradually leads to abnormal fibroblastic proliferation in the adjacent joint capsule.40 Moreover, in patients with adhesive capsulitis, synovial tissue has been found to have increased expression of inflammatory cytokines.20 Treatment with an intra-articular steroid injection is intended to control pain by reducing inflammation and prevent progression of capsular fibrosis. Several systematic reviews with meta-analyses have reported that intra-articular steroid injections are effective in patients with adhesive capsulitis for relieving pain, improving functional performance, and increasing ROMs.23,25 Our meta-analysis expands on these findings to suggest that both anterior and posterior glenohumeral joint approaches for intra-articular steroid injections are effective at reducing pain 12 weeks. In addition, when hydrodilatation is performed along with steroid injection, the anterior approach may provide greater pain reduction at 12 weeks. Since the primary pathologic sites in adhesive capsulitis are thought to involve the rotator interval, anterior/axillary joint capsule, and coracohumeral ligament,40–42 the anterior approach, directly targeting some of these structures might have contributed to more pain reduction than the posterior approach. This theory is supported by a recent RCT (excluded from the present review) that demonstrated greater improvements in VAS and shoulder ROM (in abduction, external rotation and internal rotation) when patients received combined injections to the rotator interval and inferior glenohumeral capsule compared with posterior glenohumeral joint injections.35 Previous studies have also demonstrated benefit from subacromial–subdeltoid bursa injection in patients with adhesive capsulitis.31,43 It is therefore possible that steroid injection into the anterior rotator interval, especially with hydrodilatation, may result in greater improvements in pain due to the spread of medication into the adjacent subacromial–subdeltoid bursa.

Range of Motion

Given the progressive nature of adhesive capsulitis, it is crucial to preserve ROM in the early stage of disease. Usually, the coracohumeral ligament is the first structure to be affected. Thickening and shortening of this ligament, which forms the roof of the rotator interval,44 can result in restriction of external rotation.45,46 Furthermore, thickening and contracture of the rotator interval capsule and ligamentous structures may lead to multidirectional limitation of shoulder motion as demonstrated in a cadaveric study which found that sectioning of these structures improved passive ROM in flexion, extension, external rotation, and adduction.47 Given these observations, the anterior approach has been postulated to offer greater improvements in ROM. The results of our meta-analysis partially support this hypothesis given greater improvements in external rotation and abduction with the anterior approach.

Restriction in internal rotation, on the other hand, is thought to be limited by both anterior structures (coracohumeral ligament)48 and posterior structures (posterior capsule tightness).49,50 This may explain why there was no significant difference between the 2 approaches with internal rotation. Further studies are needed to elucidate mechanisms on how different injection techniques selectively affect specific ROM measurements.

Accuracy

Targeting the rotator interval may be more technically challenging than the posterior glenohumeral joint due to its smaller space and volume.51,52 Moreover, a previous cadaver study showed that the anatomy of the rotator interval can vary among patients,44 which may explain lower accuracy of the anterior injection in one study27 included in our meta-analysis. Interestingly, the accuracy of the anterior approach was significantly higher using anatomic landmarks compared with the posterior approach in another study included in our meta-analysis.34 These conflicting results may suggest that the accuracy may be operator dependent. Our meta-analysis of 3 studies27,29,34 showed that accuracy for both anterior and posterior approaches was 87% with overlapping confidence intervals. Similar accuracy between the 2 approaches may also suggest that there may be pathologic structures which are preferentially targeted through the anterior approach compared with the posterior approach. Future trials are warranted to investigate whether accuracy is associated with clinical outcomes, if there is a difference in accuracy using ultrasound and anatomic landmark guidance, and if there is a difference in accuracy between experienced and inexperienced operators.

Clinical Implication

Our meta-analysis supports the utilization of either anterior or posterior glenohumeral joint steroid injections in the treatment of adhesive capsulitis for pain control. However, the anterior approach may be more beneficial in restoring shoulder external rotation and abduction. The subgroup analysis also revealed that when combining steroid injection and hydrodilatation, the anterior rotator interval approach may offer greater pain reduction than the posterior approach. Based on the included studies, there was no major adverse event reported with either approach. Other potential benefits of the anterior approach include better visualization through rotator interval than posterior glenohumeral recess (especially in obese patients), as well as the ability to monitor facial expressions of pain or discomfort with the anterior approach.28,53 Positioning time was also significantly faster with the anterior approach.29

Limitations

This study has several limitations. First, steroid type and doses as well as rehabilitation protocols following injections were not uniform across the studies. However, we performed random-effects meta-analyses and subgroup analyses to address such heterogeneity. Additional studies assessing higher steroid volume may be necessary given that none of the included studies used more than 40 mg of triamcinolone or methylprednisolone. Second, the results of several secondary outcomes or subgroup analyses derived from only 2 studies, reflecting the need for future trials to substantiate these findings. Third, most studies included in our analysis were published in Asian countries, including South Korea, China, and Taiwan (n = 5). It is unknown whether the results of this meta-analysis may be generalizable to other populations. Finally, the follow-up times of the included studies were limited to 12 weeks. Owing to the progressive nature of adhesive capsulitis, further studies are needed to assess if there is any difference between the anterior and posterior approaches during long-term follow-ups, at least 6 months, if not one year, or following multiple injections.

CONCLUSIONS

There was no significant difference in terms of pain control between the anterior approach and posterior glenohumeral joint approaches. However, the anterior approach may be more beneficial in restoring shoulder external rotation and abduction at 12 weeks. In addition, steroid injection combined with hydrodilatation may offer greater pain reduction when performed with the anterior rotator interval approach at 12 weeks.

ACKNOWLEDGMENTS

The authors thank Anne Fladger at Countway library, Harvard Medical School for helping us with building search terms and identifying relevant studies for this review.

References

1. Bunker TD. Frozen shoulder: unravelling the enigma. Ann R Coll Surg Engl. 1997;79:210–213.

2. Walker-Bone K, Palmer KT, Reading I, et al. Prevalence and impact of musculoskeletal disorders of the upper limb in the general population. Arthritis Rheum. 2004;51:642–651.

3. Sarasua SM, Floyd S, Bridges WC, et al. The epidemiology and etiology of adhesive capsulitis in the U.S. Medicare population. BMC Musculoskelet Disord. 2021;22:828.

4. Kelley MJ, Shaffer MA, Kuhn JE, et al. Shoulder pain and mobility deficits: adhesive capsulitis. J Orthop Sports Phys Ther. 2013;43:A1–A31.

5. Juel NG, Brox JI, Brunborg C, et al. Very high prevalence of frozen shoulder in patients with type 1 diabetes of ≥45 years’ duration: the dialong shoulder study. Arch Phys Med Rehabil. 2017;98:1551–1559.

6. de la Serna D, Navarro-Ledesma S, Alayón F, et al. A comprehensive view of frozen shoulder: a mystery syndrome. Front Med (Lausanne). 2021;8:663703.

7. Schiefer M, Teixeira PFS, Fontenelle C, et al. Prevalence of hypothyroidism in patients with frozen shoulder. J Shoulder Elbow Surg. 2017;26:49–55.

8. Wang JY, Liaw CK, Huang CC, et al. Hyperlipidemia is a risk factor of adhesive capsulitis: real-world evidence using the Taiwanese national health insurance research database. Orthop J Sports Med. 2021;9:2325967120986808.

9. Li W, Lu N, Xu H, et al. Case control study of risk factors for frozen shoulder in China. Int J Rheum Dis. 2015;18:508–513.

10. Itoi E, Arce G, Bain GI, et al. Shoulder stiffness: current concepts and concerns. Arthroscopy. 2016;32:1402–1414.

11. Kingston K, Curry EJ, Galvin JW, et al. Shoulder adhesive capsulitis: epidemiology and predictors of surgery. J Shoulder Elbow Surg. 2018;27:1437–1443.

12. Pietrzak M. Adhesive capsulitis: an age related symptom of metabolic syndrome and chronic low-grade inflammation? Med Hypotheses. 2016;88:12–17.

13. Hsu JE, Anakwenze OA, Warrender WJ, et al. Current review of adhesive capsulitis. J Shoulder Elbow Surg. 2011;20:502–514.

14. Le HV, Lee SJ, Nazarian A, et al. Adhesive capsulitis of the shoulder: review of pathophysiology and current clinical treatments. Shoulder Elbow. 2017;9:75–84.

15. Manske RC, Prohaska D. Diagnosis and management of adhesive capsulitis. Curr Rev Musculoskelet Med. 2008;1:180–189.

16. Neviaser AS, Neviaser RJ. Adhesive capsulitis of the shoulder. J Am Acad Orthop Surg. 2011;19:536–542.

17. Millar NL, Meakins A, Struyf F, et al. Frozen shoulder. Nat Rev Dis Primers. 2022;8:59.

18. Ogilvie-Harris DJ, Biggs DJ, Fitsialos DP, et al. The resistant frozen shoulder. Manipulation versus arthroscopic release. Clin Orthop Relat Res. 1995;319:238–248.

19. Wiley AM. Arthroscopic appearance of frozen shoulder. Arthroscopy. 1991;7:138–143.

20. Tamai K, Akutsu M, Yano Y. Primary frozen shoulder: brief review of pathology and imaging abnormalities. J Orthop Sci. 2014;19:1–5.

21. Petchprapa CN, Beltran LS, Jazrawi LM, et al. The rotator interval: a review of anatomy, function, and normal and abnormal MRI appearance. AJR Am J Roentgenol. 2010;195:567–576.

22. Yip M, Francis AM, Roberts T, et al. Treatment of adhesive capsulitis of the shoulder: a critical analysis review. JBJS Rev. 2018;6:e5.

23. Challoumas D, Biddle M, McLean M, et al. Comparison of treatments for frozen shoulder: a systematic review and meta-analysis. JAMA Netw Open. 2020;3:e2029581.

24. Catapano M, Mittal N, Adamich J, et al. Hydrodilatation with corticosteroid for the treatment of adhesive capsulitis: a systematic review. PM R. 2018;10:623–635.

25. Sun Y, Zhang P, Liu S, et al. Intra-articular steroid injection for frozen shoulder: a systematic review and meta-analysis of randomized controlled trials with trial sequential analysis. Am J Sports Med. 2017;45:2171–2179.

26. Prestgaard T, Wormgoor MEA, Haugen S, et al. Ultrasound-guided intra-articular and rotator interval corticosteroid injections in adhesive capsulitis of the shoulder: a double-blind, sham-controlled randomized study. PAIN. 2015;156:1683–1691.

27. Cho CH, Kim DH, Kim DH, et al. Comparative efficacy of rotator interval versus posterior capsule approach intraarticular corticosteroid injections for primary frozen shoulder: a single-blind, randomized trial. Pain Physician. 2022;25:313–321.

28. Elnady B, Rageh EM, Hussein MS, et al. In shoulder adhesive capsulitis, ultrasound-guided anterior hydrodilatation in rotator interval is more effective than posterior approach: a randomized controlled study. Clin Rheumatol. 2020;39:3805–3814.

29. Kim DY, Lee SS, Nomkhondorj O, et al. Comparison between anterior and posterior approaches for ultrasound-guided glenohumeral steroid injection in primary adhesive capsulitis: a randomized controlled trial. J Clin Rheumatol. 2017;23:51–57.

30. Deng Z, Li Z, Li X, et al. Comparison of outcomes of two different corticosteroid injection approaches for primary frozen shoulder: a randomized controlled study. J Rehabil Med. 2023;55:jrm00361.

31. Sun Y, Liu S, Chen S, et al. The effect of corticosteroid injection into rotator interval for early frozen shoulder: a randomized controlled trial. Am J Sports Med. 2018;46:663–670.

32. Wang JC, Tsai PY, Hsu PC, et al. Ultrasound-guided hydrodilatation with triamcinolone acetonide for adhesive capsulitis: a randomized controlled trial comparing the posterior glenohumeral recess and the rotator cuff interval approaches. Front Pharmacol. 2021;12:686139.

33. Page MJ, McKenzie JE, Bossuyt PM, et al. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews. BMJ. 2021;372:n71.

34. Rijs Z, de Groot PCJ, Zwitser EW, et al. Is the anterior injection approach without ultrasound guidance superior to the posterior approach for adhesive capsulitis of the shoulder? A sequential, prospective trial. Clin Orthop Relat Res. 2021;479:2483–2489.

35. Lee SH, Choi HH, Chang MC. Comparison between corticosteroid injection into coracohumeral ligament and inferior glenohumeral capsule and corticosteroid injection into posterior glenohumeral recess in adhesive capsulitis: a prospective randomized trial. Pain Physician. 2022;25:E787–E793.

36. Sterne JAC, Savović J, Page MJ, et al. RoB 2: a revised tool for assessing risk of bias in randomised trials. BMJ. 2019;366:l4898.

37. Wells GA, Shea B, O’Connell D, et al. The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomised studies in meta-analyses; 2000. Available at: http://www.ohri.ca/programs/clinical_epidemiology/oxford.asp. Accessed April 13, 2023.

38. Higgins JP, Thompson SG, Deeks JJ, et al. Measuring inconsistency in meta-analyses. BMJ. 2003;327:557–560.

39. Takeshima N, Sozu T, Tajika A, et al. Which is more generalizable, powerful and interpretable in meta-analyses, mean difference or standardized mean difference? BMC Med Res Methodol. 2014;14:30.

40. Hand GC, Athanasou NA, Matthews T, et al. The pathology of frozen shoulder. J Bone Joint Surg Br. 2007;89:928–932.

41. Won KS, Kim DH, Sung DH, et al. Clinical correlation of metabolic parameters on (18)F-FDG PET/CT in idiopathic frozen shoulder. Ann Nucl Med. 2017;31:211–217.

42. Bunker TD, Anthony PP. The pathology of frozen shoulder. A Dupuytren-like disease. J Bone Joint Surg Br. 1995;77:677–683.

43. Shin SJ, Lee SY. Efficacies of corticosteroid injection at different sites of the shoulder for the treatment of adhesive capsulitis. J Shoulder Elbow Surg. 2013;22:521–527.

44. Yang HF, Tang KL, Chen W, et al. An anatomic and histologic study of the coracohumeral ligament. J Shoulder Elbow Surg. 2009;18:305–310.

45. Lee SY, Park J, Song SW. Correlation of MR arthrographic findings and range of shoulder motions in patients with frozen shoulder. AJR Am J Roentgenol. 2012;198:173–179.

46. Neer CS, Satterlee CC, Dalsey RM, et al. The anatomy and potential effects of contracture of the coracohumeral ligament. Clin Orthop Relat Res. 1992;280:182–185.

47. Harryman DT II, Sidles JA, Harris SL, et al. The role of the rotator interval capsule in passive motion and stability of the shoulder. J Bone Joint Surg Am. 1992;74:53–66.

48. Koide M, Hamada J, Hagiwara Y, et al. A thickened coracohumeral ligament and superomedial capsule limit internal rotation of the shoulder joint: report of three cases. Case Rep Orthop. 2016;2016:9384974.

49. Burkhart SS, Morgan CD, Kibler WB. The disabled throwing shoulder: spectrum of pathology Part I: pathoanatomy and biomechanics. Arthroscopy. 2003;19:404–420.

50. Tehranzadeh AD, Fronek J, Resnick D. Posterior capsular fibrosis in professional baseball pitchers: case series of MR arthrographic findings in six patients with glenohumeral internal rotational deficit. Clin Imaging. 2007;31:343–348.

51. Juel NG, Oland G, Kvalheim S, et al. Adhesive capsulitis: one sonographic-guided injection of 20 mg triamcinolon into the rotator interval. Rheumatol Int. 2013;33:1547–1553.

52. Pape JL, Boudier-Revéret M, Brismée JM, et al. Accuracy of unguided and ultrasound guided Coracohumeral ligament infiltrations—a feasibility cadaveric case series. BMC Musculoskelet Disord. 2020;21:136.

53. Yoong P, Duffy S, McKean D, et al. Targeted ultrasound-guided hydrodilatation via the rotator interval for adhesive capsulitis. Skeletal Radiol. 2015;44:703–708.

Leave a Reply

Your email address will not be published. Required fields are marked *