Goulart LT, Matsunaga FT, Belloti JC, Netto NA, Paim TS, Tamaoki MJS. Effectiveness of sub-acromial injections in rotator cuff injuries: A systematic review and meta-analysis. World J Orthop 2025; 16(2): 102856 [DOI: 10.5312/wjo.v16.i2.102856]
Corresponding Author of This Article
Luana Tossolini Goulart, Department of Orthopedics and Trauma, Federal University of São Paulo-Paulista School of Medicine, 740 Botucatu Street, São Paulo 04023-062, Brazil. goulart.luana@unifesp.br
Research Domain of This Article
Orthopedics
Article-Type of This Article
Meta-Analysis
Open-Access Policy of This Article
This article is an open-access article which was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/
Luana Tossolini Goulart, Fabio Teruo Matsunaga, Joao Carlos Belloti, Nicola Archetti Netto, Thays Sellan Paim, Marcel Jun Sugawara Tamaoki, Department of Orthopedics and Trauma, Federal University of São Paulo-Paulista School of Medicine, São Paulo 04023-062, Brazil
Author contributions: Goulart LT and Tamaoki MJS devised the project and collected data in the literature on the content covered; Tamaoki MJS conceived the study and was in charge of overall supervision; Matsunaga FT structured the manuscript according to scientific norms; Goulart LT and Paim TS contributed to the design and wrote the manuscript, evaluated the data, and planned the meta-analyses; Belloti JC and Netto NA reviewed the study proposal and contributed to implementation of the research.
Conflict-of-interest statement: All the authors report no relevant conflicts of interest for this article.
PRISMA 2009 Checklist statement: The authors have read the PRISMA 2009 Checklist, and the manuscript was prepared and revised according to the PRISMA 2009 Checklist.
Open Access: This article is an open-access article that was selected by an in-house editor and fully peer-reviewed by external reviewers. It is distributed in accordance with the Creative Commons Attribution NonCommercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: https://creativecommons.org/Licenses/by-nc/4.0/
Corresponding author: Luana Tossolini Goulart, Department of Orthopedics and Trauma, Federal University of São Paulo-Paulista School of Medicine, 740 Botucatu Street, São Paulo 04023-062, Brazil. goulart.luana@unifesp.br
Received: October 31, 2024 Revised: December 20, 2024 Accepted: January 17, 2025 Published online: February 18, 2025 Processing time: 104 Days and 11.8 Hours
Abstract
BACKGROUND
Sub-acromial injections are a therapeutic option for rotator cuff injuries; however, evidence regarding the most effective drug in this context is unclear, which needs to be investigated.
AIM
To evaluate the effectiveness of various sub-acromial injections for rotator cuff injuries.
METHODS
We conducted a systematic review and pair-wise and network meta-analyses of randomized clinical trials (RCTs) comparing sub-acromial injections for rotator cuff injuries. The interventions evaluated were hyaluronic acid (HA), platelet-rich plasma (PRP), prolotherapy, and corticosteroids. The outcomes of interest were pain and functional improvement, which were evaluated with standardized scores. The Risk of Bias 2 tool and the Grading of Recommendations, Assessment, Development and Evaluation methodology were used to assess data quality.
RESULTS
Twenty RCTs, comprising 1479 participants, were included. In the short term, HA achieved the best outcomes [pain mean difference (MD) = -1.48, 95% confidence interval (CI) -2.37 to -0.59; function MD = 10.18, 95%CI: 4.96-15.41]. In the medium term, HA, PRP, HA + PRP, and corticosteroids were not superior to placebo in improving pain. Based on function, HA + PRP was superior to placebo, corticosteroids, and PRP (MD = 26.72; 95%CI: 8.02-45.41). In the long term, HA, PRP, and corticosteroids were not superior to placebo in reducing pain. However, based on function, HA + PRP, PRP, and HA were superior to placebo, and HA + PRP had the best result (MD = 36.64; 95%CI: 31.66-33.62).
CONCLUSION
HA provides satisfactory short-term results, while HA with PRP demonstrates functional improvement in the medium and long terms. However, no intervention maintained the pain-relief effect on > 3-month follow-up.
Core Tip: This study investigates, through network meta-analyses, the effectiveness of different sub-acromial injections for rotator cuff injuries. The main findings include that hyaluronic acid (HA) offers the best short-term pain relief, while HA combined with platelet-rich plasma shows the most significant functional improvement in the medium and long terms. None of the interventions sustained pain relief beyond three months. These results provide critical insights for optimising therapeutic strategies for rotator cuff injuries.
Citation: Goulart LT, Matsunaga FT, Belloti JC, Netto NA, Paim TS, Tamaoki MJS. Effectiveness of sub-acromial injections in rotator cuff injuries: A systematic review and meta-analysis. World J Orthop 2025; 16(2): 102856
Shoulder pain is the third most common complaint in orthopaedic care, often caused by rotator cuff injuries[1,2]. These injuries affect approximately 25%-40% of individuals of > 60 years of age and > 50% of individuals > 80 years of age[3,4]. Rotator cuff injuries may involve inflammatory processes and fibrotic changes as well as a complete rupture of the tendons[5,6].
No consensus regarding the aetiopathogenesis of rotator cuff injuries exists; nevertheless, both intrinsic factors, related to degenerative processes, and extrinsic factors, associated with sub-acromial impingement, may be involved[7,8]. Many treatment options are available and non-surgical approaches are often recommended for eligible patients. Injections are increasingly used as they are generally perceived as safe, accessible, and non-invasive when compared to surgeries[9]. The most commonly used substances are corticosteroids, hyaluronic acid (HA), platelet-rich plasma (PRP), and prolotherapy.
Corticosteroids, by modifying the mechanisms of protein synthesis, have anti-inflammatory effects, mediated by the inhibition of the migration of polymorphonuclear cells and macrophages and the suppression of the immunological response of lymphocytes[10]. HA contributes to the lubrication, hydration, and nutrition of the joint tissue, helping stimulate the proliferation of chondrocytes, increase the production of type-2 collagen, and reduce its degradation rate[11]. The use of HA in tendinopathies is based on anti-inflammatory, analgesic, or pro-inflammatory and potentially regenerative properties, depending on the weight of the molecule[12]. PRP is a preparation with a high concentration of platelets, obtained by the centrifugation of autologous blood[13]. In addition to acting in the haemostatic cascade, platelets can induce an inflammatory response mediated by cytokines and growth factors[14]. In addition, PRP can support tissue repair with the plasma proteins present in this preparation, especially fibronectin, prothrombin, and fibrinogen[15]. Prolotherapy involves the application of irritating or sclerosing substances to the areas of musculoskeletal tissue, such as joints, tendons, and ligaments, with the intention of promoting tissue regeneration and growth[16]. The most used agent is dextrose (d-glucose), which is a natural form of glucose at concentrations > 10%[17,18].
Although injections are widely used in clinical practice, the supporting evidence remains limited. No established consensus on the effectiveness of the available medications or orthobiologics or on the most appropriate one exists.
The last systematic review[19] on this topic was published in 2022 and included only patients with shoulder pain lasting for < 3 months, diagnosed by magnetic resonance imaging (MRI), and receiving injections with at least two medications or biological agents (HA, PRP, corticosteroids, and/or prolotherapy). The reported outcomes of interest were pain scores but not functional gains. This previous study excluded patients with chronic conditions and those diagnosed by ultrasound (US), as well as placebo-controlled trials. To address these limitations, the present systematic review and meta-analysis aimed to evaluate and compare the effectiveness of sub-acromial injections of medications and biological products for the treatment of rotator cuff injuries, based on pain and function improvement scores, as evaluated in randomized clinical trials (RCTs).
MATERIALS AND METHODS
Study guidelines and registration
This systematic review was performed based on the Cochrane Manual for Systematic Reviews and complied with the PRISMA guidelines[20]. The study was registered on the International Systematic Review Registration Platform (PROSPERO)-CRD42020199292-and a protocol for conducting this systematic review was published in BMJ Open on November 2, 2022[21].
Eligibility criteria
This review included RCTs that compared different sub-acromial injections (HA, PRP, prolotherapy, corticosteroids, and/or placebo) for the treatment of symptomatic rotator cuff injuries, diagnosed by MRI or US, in adults of > 18 years of age. Studies with other methodological designs and those that used modalities other than shoulder injections as a control group (such as physiotherapy, acupuncture, exercises, or surgery) were excluded. Studies that recruited patients with the following characteristics were also excluded: Clinical diagnosis of a rotator cuff injury, total rotator cuff tear, or other shoulder diseases (adhesive capsulitis, shoulder osteoarthritis, shoulder instability, and fractures); history of surgery; receiving intra-articular injections; and/or treated with substances other than PRP, HA, prolotherapy, corticosteroids, or placebo.
Outcome
The primary outcome was the improvement in shoulder pain after sub-acromial injection measured by the visual analogue scale (VAS)[22] and functional gain using standardized scores: The Constant-Murley Score (CMS)[23]; shoulder pain disability index (SPADI)[24]; American Shoulder and Elbow Surgeons’ Society Standardized Shoulder Assessment score (ASES)[25]; Simple Shoulder Test (SST)[26]; Western Ontario Rotator Cuff index (WORC)[27]; Disabilities of the arm, shoulder and hand[28]; and University of California at Los Angeles Shoulder Score[29]. The results were evaluated in the short, medium, and long terms, which were defined as < 3 months, 3–12 months, and > 12 months, respectively. The secondary outcomes were adverse events and radiological findings observed after the injections.
Search strategy
Publications that met the eligibility criteria for this review were searched in MEDLINE/PubMed, Cochrane Library, LILACS, and EMBASE. The search was expanded to reference lists, conference annals, abstracts, and academic papers (including theses and dissertations), published up to December 2023. No restrictions were imposed on the year or language of publication. The search strategies for each platform are presented in Table 1. The search strategy was modified from that described in the initial protocol to help capture the maximum number of studies reporting on the outcomes and populations of interest.
Table 1 Search strategies presented to the main databases.
Databases
Search strategy
Medline/PubMed
(“Rotator cuff” OR shoulder impingement syndrome OR subacromial bursitis OR subacromial impingement OR rotator cuff injuries OR rotator cuff tendinosis OR shoulder pain OR supraspinatus OR infraspinatus OR subscapular OR teres minor) AND (hyaluronic acid OR hyaluronic acid injection OR sodium hyaluronate OR hyaluronate sodium OR viscosupplement platelet rich plasma injection OR platelet rich plasma OR PRP OR corticoid injection OR glucocorticoid OR corticosteroid injection OR prolotherapy OR dextrose injection) NOT (arthroscopy OR surgery OR adhesive capsulitis OR frozen shoulder)
(“rotator cuff” OR “shoulder impingement syndrome” OR “subacromial bursitis” OR “subacromial impingement” OR “rotator cuff injuries” OR “rotator cuff tendinosis” OR “rotator cuff lesions” OR “shoulder pain” OR “supraspinatus” OR “infraspinatus” OR “subscapular” OR “teres minor”) AND (“hyaluronic acid” OR “viscosupplement” OR “hyaluronate sodium” OR “PRP” OR “platelet rich plasma” OR “corticoid” OR “corticosteroid” OR “glucocorticoid” OR “prolotherapy” OR “dextrose injection”) NOT (“arthroscopy” OR “surgery” OR “adhesive capsulitis” OR “frozen shoulder”)
LILACS
TW*: (“rotator cuff” or “shoulder impingement syndrome” or " “rotator cuff injuries" or “rotator cuff lesions” or “rotator cuff tendinosis” or “subacromial bursitis” or “subacromial impingement”) AND (“hyaluronic acid” or “viscosupplement” or “hyaluronate sodium” or “PRP” or “platelet rich plasma” or “corticoid” or “corticosteroid injection” or “glucocorticoid” or “prolotherapy” or “dextrose injection”) AND NOT (“arthroscopy” OR “surgery” OR “adhesive capsulitis” OR “frozen shoulder”)
Study selection
Two authors (Goulart LT and Paim TS) independently screened studies based on the eligibility criteria and identified and removed duplicates. The titles and abstracts of eligible articles were read by two researchers. Any disagreements were resolved between the team members (Tamaoki MJS or Matsunaga FT). Full-text reading of short-listed studies was performed by both researchers, at which point studies deemed ineligible were excluded. A Microsoft Excel spreadsheet (Microsoft Corporation, Redmond, WA, United States) was maintained by the researchers, recording decisions on studies with brief rationales. If additional clarification was required, study authors were contacted by email; if no reply was received, eligible studies were classified as having ‘missing information’. The details of the study selection procedure are shown in the PRISMA flowchart in Figure 1.
Figure 1
PRISMA flowchart shows the number of articles (n) in the different selection stages of the selection process.
Data extraction
The data were extracted using Microsoft Excel. Two researchers (Goulart LT and Paim TS) extracted the following information: Title, author names, year and country of publication, study design, interventions performed, follow-up time, sample size, rotator cuff injury details, injection agent (HA, PRP, corticosteroids, and prolotherapy), number of injections, use of US guidance, outcomes, and outcome measures. Furthermore, the location of the injections (sub-acromial, bursal, or intra-tendinous) was tabulated. The patients were sub-divided into two groups according to the diagnosis: Partial rotator cuff tear and rotator cuff tendinopathy. In addition, estimates of the number of patients included, mean pain scores, and the corresponding standard deviations, were obtained. For dichotomous outcomes (adverse effects), the number of patients in each group and frequency of events were recorded. Data were extracted based on an intention-to-treat analysis.
Assessment of the methodological quality of studies
The risk of bias of the selected studies was analyzed by two researchers (Goulart LT and Paim TS) using the Risk of Bias 2 tool developed by Cochrane[30]. The following five domains of bias were assessed: (1) Due to the randomisation process; (2) Due to deviations from intended interventions; (3) Due to missing outcome data; (4) Due to outcome measurement; and (5) Selection of the reported result. Each of these domains can be classified as at a ‘low’ and ‘high’ risk of bias or as showing ‘some concerns’ (indicating a lack of information or uncertainty regarding the potential for bias). Disagreements between the researchers regarding the risk of bias in each domain were resolved by a third researcher (Matsunaga FT).
Quality of evidence and strength of recommendation
The quality and degree of recommendation of the evidence for each outcome were assessed by two researchers (Goulart LT and Paim TS) using the Grading of Recommendations, Assessment, Development and Evaluation (GRADE)[31]. To use this tool, five domains that may reduce the quality of evidence were considered: Study bias, inconsistency, imprecision, indirect evidence, and publication bias. This system classifies the quality of evidence as high, moderate, low, or very low. Furthermore, it classifies a recommendation as strong or weak (also termed ‘conditional’).
Statistical analysis
The number of patients and mean and standard deviation of pain reduction and functional improvement scores were used in the meta-analysis. The particularities of each measurement score for pain and functional outcomes were considered and the results were expressed as mean difference (MD) with 95% confidence interval (CI). All statistical analyses were performed using the ‘Netmeta’ package (version 2.8-0), implemented in R language (R Foundation for Statistical Computing, Vienna, Austria), with P values of < 0.05 considered statistically significant. The statistical software used was different from that initially described in the protocol to enable a network meta-analysis. The unit of randomisation for the included studies was individual patients, not shoulders.
Pair-wise and network meta-analyses were performed for each outcome according to pre-determined evaluation periods. Random-effects network meta-analyses were performed using a frequentist approach along with pairwise meta-analyses to explore the transitivity assumption and determine the strength of the evidence. Regarding sample loss, an assessment was developed based on intention-to-treat. The P-score and surface under the cumulative ranking curve (SUCRA) were used to rank all treatments, mainly considering point estimates and precision[32].
The heterogeneity of the studies was evaluated by Cochran's Q test and Higgins and Thompson's I2 consistency index. Heterogeneity was assessed within and between studies. Consistency, defined as the difference between the direct and indirect evidence, was assessed using the split node method.
The primary studies were subdivided into the categories of partial tears and rotator cuff tendinopathy, and data on the diagnostic method, application technique (guided or not guided by US), application site (bursal or intra-tendinous), number of injections, and co-interventions were extracted, for analysing sub-groups based on these characteristics. Although we acknowledge that heterogeneity may limit the generalisability of the results, we chose to conduct a sub-group analysis based on the outcome-assessment period to verify the consistency of the results. However, following statistical principles, no further sub-group analyses were performed to avoid data dredging, i.e., conducting excessive analysis that may generate spurious and non-representative results of clinical reality. In this systematic review, a sensitivity analysis was performed to investigate the impact of including studies with a high risk of bias on the estimated effect, which showed no significant impact.
Patient and public involvement
The participants did not participate directly in the study. As this was a systematic review, all information used was obtained from previously published studies.
RESULTS
Characteristics of the studies
Twenty RCTs (2004-2022), comprising a total of 1479 (56.52% women) participants mostly in the fifth or sixth decade of life, were included in this systematic review.
Akgün et al[33], Alvarez et al[34], Barreto et al[35], Lin et al[36], Chang et al[37], Kim et al[38], Cole et al[39], Hsieh et al[40], Damjanov et al[41], Lin et al[42], and Meloni et al[43] included participants with rotator cuff tendinopathy. Cai et al[44], Chou et al[45], Kesikburun et al[46], Moghtaderi et al[47], Shams et al[48], Schwitzguebel et al[49], Kwong et al[50], Dadgostar et al[51], and Sari et al[52] included participants with partial-thickness rotator cuff tears.
Only one study was published in Portuguese, in which the present authors are fluent, allowing to evaluate its content without any need for translation[35]. The other studies were published in English. All included RCTs used complementary examinations (US and/or MRI) to diagnose rotator cuff injuries and 15 studies used US to guide shoulder injections. The main characteristics of the included studies are presented in Tables 2 and 3.
Table 2 The main characteristics of the included studies for partial-thickness rotator cuff tears.
Of the 20 studies included in this systematic review, only three, covering three comparisons (9%) were deemed at low risk of bias. Five studies, representing seven comparisons (22%), were classified as having a high risk of bias, and the other 12 studies, which included 22 comparisons (69%), were evaluated as showing ‘some concerns’. The general and detailed risk of bias for each study is presented in Figures 2 and 3, respectively. The quality of evidence, employing GRADE, for each outcome is shown in Table 4 (summary of findings). For all outcomes, the quality of evidence ranged from very low to moderate in most cases.
Short-term effectiveness of sub-acromial injections
Fifteen studies evaluated short-term pain by comparing six treatments: HA, corticosteroids, HA + PRP, PRP, prolotherapy, and placebo. The network meta-analysis showed a statistically significant reduction in pain only with HA (MD = -1.48, 95%CI: -2.37 to -0.59) and corticosteroids (MD = -0.92, 95%CI: -1.73 to -0.12), when compared to placebo. Table 5 (league table-short term VAS) shows the results of all comparisons between treatments. Pain reduction was greater for those who received HA injection than for those who received prolotherapy (MD = -1.43, 95%CI: -2.66 to -0.20). The minimal clinically important difference (MCID) reported for the non-surgical treatment of rotator cuff injuries was 1.4 points on the VAS[53], suggesting that HA was statistically and clinically superior to placebo and prolotherapy, while corticosteroids, although statistically superior to placebo, did not achieve MCID in the short term. Supplementary Figure 1 demonstrates the probabilities of classifying the best treatment for pain in the short term, with the probability of HA being the first option at 57.9%.
Table 5 League Table with the results of the network meta-analysis for pain improvement Visual Analogue Scale in the short term.
Hyaluronic acid
HA + PRP
Corticosteroids
PRP
Prolotherapy
Placebo
Network meta-analysis-MD
-0.28 [-1.86; 1.30]
-0.56 [-1.54; 0.42]
-0.28 [-1.90; 1.35]
-0.92 [-2.00; 0.16]
-0.64 [-2.22; 0.93]
-0.37 [-1.25; 0.52]
-1.43 [-2.66; -0.20]
-1.15 [-2.90; 0.60]
-0.88 [-1.99; 0.24]
-0.51 [-1.67; 0.65]
-1.48 [-2.37; -0.59]
-1.20 [-2.74; 0.34]
-0.92 [-1.73; -0.12]
-0.56 [-1.45; 0.33]
-0.05 [-0.95; 0.86]
Regarding function assessed by CMS, six studies investigated HA, HA + PRP, corticosteroids, PRP, and placebo, with all options being superior to the placebo, except for corticosteroids (Table 6; league table-CMS short-term). Considering the MCID reported for the treatment of rotator cuff tendinopathy of 10 points for CMS[54,55], both HA and HA + PRP showed clinical superiority over the placebo for short-term functional improvements. Despite being statistically superior to placebo, PRP did not achieve MCID. Supplementary Figure 2 demonstrates the probabilities of classifying the best treatment for function in the short term, based on the CMS, with the probability of HA + PRP being the first option at 46.3%.
Table 6 League Table with the results of the network meta-analysis for function improvement Constant Murley Score in the short term.
HA + PRP
Hyaluronic acid
PRP
Corticosteroids
Placebo
Network meta-analysis-MD
0.53 [-6.22; 7.28]
3.15 [-3.89; 10.18]
2.62 [-3.89; 9.12]
5.84 [-4.07; 15.75]
5.31 [-4.00;14.62]
2.69 [-5.51; 10.89]
10.7 [4.01; 17.42]
10.18 [4.96; 15.41]
7.57 [1.31; 13.82]
4.87 [-3.69; 13.43]
For ASES, six studies evaluated the use of HA, HA + PRP, corticosteroids, PRP, prolotherapy, and placebo, with none of these substances being superior to the placebo. Pooled results, shown in Table 7 (league table-ASES short term), revealed the superiority of the HA + PRP combination over PRP (MD = 10.61, 95%CI: 0.11-21.11), which was clinically insignificant, given the MCID in the range of 12-17 points[56,57]. Supplementary Figure 3 demonstrates the probabilities of classifying the best treatment for function in the short term, based on the ASES, with the probability of HA + PRP being the first option at 78.4%.
Table 7 League Table with the results of the network meta-analysis for function improvement American Shoulder and Elbow Surgeons Standardized Shoulder Assessment in the short term.
HA + PRP
Corticosteroids
Hyaluronic acid
Placebo
Prolotherapy
PRP
Network meta-analysis-MD
6.15 [-5.00; 17.31]
7.09 [-3.79; 17.97]
0.93 [-7.47; 9.34]
8.97 [-1.61; 19.55]
2.82 [-4.94; 10.57]
1.88 [-7.18; 10.94]
9.95 [-4.36; 24.25]
3.79 [-7.41; 15.00]
2.86 [-10.01; 15.73]
0.98 [-10.38; 12.34]
10.61 [0.11; 21.11]
4.45 [-2.37; 11.28]
3.52 [-5.23; 12.26]
1.64 [-6.00; 9.27]
0.66 [-10.44; 11.76]
Medium-term outcomes
Seven studies investigated medium-term pain improvement using four treatments (HA, corticosteroids, PRP, and placebo). None of the options was statistically or clinically superior to the placebo. As shown in Table 8 (league table-medium term VAS), PRP was statistically and clinically superior to corticosteroids (MD = -1.56, 95%CI: -2.88 to -0.25). Data presented in Supplementary Figure 4 SUCRA-medium-term VAS) represent the probabilities of classifying the best treatment for pain in the medium term, with the probability of HA being the first option at 58.3%.
Table 8 League Table with the results of the network meta-analysis for pain improvement Visual Analogue Scale in the medium term.
Hyaluronic acid
PRP
Placebo
Corticosteroids
Network meta-analysis-MD
-0.24 [-2.21; 1.74]
-1.50 [-3.48; 0.48]
-1.27 [-2.38; -0.16]
-1.80 [-4.01; 0.41]
-1.56 [-2.88; -0.25]
-0.30 [-1.51; 0.91]
Four studies evaluated function through the CMS in the medium-term, covering five injection options: HA, HA + PRP, corticosteroids, PRP, and placebo. As shown in Table 9 (league table-CMS medium-term), no statistically significant difference was observed between the treatments. A network meta-analysis revealed that the probability of HA + PRP being the top treatment option was 76.2% (Supplementary Figure 5, SUCRA-CMS medium term).
Table 9 League Table with the results of the network meta-analysis for function improvement Constant Murley Score in the medium term.
HA + PRP
PRP
Hyaluronic acid
Corticosteroids
Placebo
Network meta-analysis-MD
13.81 [-12.54; 40.16]
15.87 [-12.23; 43.97]
2.06 [-24.30; 28.41]
21.42 [-16.63; 59.48]
7.61 [-19.84; 35.06]
5.55 [-32.50; 43.61]
22.73 [-3.62; 49.08]
8.92 [-11.28; 29.11]
6.86 [-19.49; 33.21]
1.30 [-32.78; 35.38]
Considering the improvement in function in the medium term, based on the ASES, five studies evaluated six treatments (HA, HA + PRP, corticosteroids, PRP, prolotherapy, and placebo). HA + PRP was statistically and clinically superior to PRP (MD = 19.09, 95%CI: 0.37-37.81), corticosteroids (MD = 26.44, 95%CI: 4.86-48.02), and placebo (MD = 26.72, 95%CI: 8.02-45.41) (Table 10; league table-ASES medium term). In the classification of probabilities, HA + PRP had a 95.2% probability of being the top treatment (Supplementary Figure 6; SUCRA-ASES medium term).
Table 10 League Table with the results of the network meta-analysis for function improvement American Shoulder and Elbow Surgeons Standardized Shoulder Assessment in the medium term.
HA + PRP
PRP
Hyaluronic acid
Prolotherapy
Corticosteroids
Placebo
Network meta-analysis-MD
19.09 [0.37; 37.81]
20.29 [-0.38; 40.76]
1.10 [-17.67; 19.87]
23.63 [-1.58; 48.84]
4.54 [-14.22; 23.30]
3.44 [-21.81; 28.69]
26.44 [4.86; 48.02]
7.35 [-6.09; 20.80]
6.25 [-15.37; 27.88]
2.81 [-16.37; 22.00]
26.72 [8.02; 45.41]
7.63 [-3.98; 19.23]
6.53 [-12.23; 25.28]
3.09 [-15.71; 21.89]
0.27 [-13.34; 13.88]
Long-term outcomes
For long-term pain assessment, four studies analysed four injection possibilities (HA, corticosteroids, PRP, and placebo). The meta-analyses revealed no statistically significant pain improvement among the treatments (Table 11; league table-long-term VAS). The probability of HA being the first treatment option was 45.9% (Supplementary Figure 7; SUCRA-long-term VAS).
Table 11 League Table with the results of the network meta-analysis for pain improvement Visual Analogue Scale in the long term.
Hyaluronic acid
PRP
Corticosteroids
Placebo
Network meta-analysis-MD
-0.48 [-5.72; 4.76]
-1.59 [-9.41; 6.23]
-1.11 [-6.91; 4.69]
-2.59 [-7.83; 2.65]
-2.11 [-5.49; 1.27]
-1.00 [-7.72; 5.71]
Only one study evaluated long-term functional outcomes using the CMS, based on four comparisons (HA, HA + PRP, PRP, and placebo). The network meta-analysis showed a statistically and clinically significant improvement in comparisons against placebo, for HA (MD = 13.17, 95%CI: 12.14-14.20), HA + PRP (MD = 32.64, 95%CI: 31.66-33.62), and PRP (MD = 24.40, 95%CI: 23.35-25.45). Among these treatments, HA + PRP (MD = 19.47, 95%CI: 18.51-20.43) and PRP (MD = 11.23, 95%CI: 10.20-12.26) were superior to HA (Table 12; league table-CMS long term). For this outcome, the classification of probabilities was not applicable because only one study included four groups.
Table 12 League Table with the results of the network meta-analysis for function improvement Constant Murley Score in the medium term.
HA + PRP
PRP
Hyraluronic acid
Placebo
Network meta-analysis-MD
8.24 [7.26; 9.22]
19.47 [18.51; 20.43]
11.23 [10.20; 12.26]
32.64 [31.66; 33.62]
24.40 [23.35; 25.45]
13.17 [12.14; 14.20]
Long-term ASES scores were evaluated in two studies that included five treatments (HA, HA + PRP, corticosteroids, PRP, and placebo). As shown in Table 13 (league table-ASES long-term), all options were clinically superior to the placebo. HA + PRP was clinically superior to PRP (MD = 13.58, 95%CI: 11.70-15.46), corticosteroids (MD = 20.88, 95%CI: 12.25-29.51), and HA (MD = 28.45, 95%CI: 26.76-30.14). Furthermore, PRP was superior to HA (MD = 14.87, 95%CI: 12.76-16.98). For this outcome, the classification of probabilities was not applicable because only two studies, precluding grouping in direct pairwise meta-analyses, were included to estimate indirect effects.
Table 13 League Table with the results of the network meta-analysis for function improvement American Shoulder and Elbow Surgeons Standardized Shoulder Assessment in the long term.
HA + PRP
Corticosteroids
Hyaluronic acid
Placebo
Prolotherapy
PRP
Network meta-analysis-MD
6.15 [-5.00; 17.31]
7.09 [-3.79; 17.97]
0.93 [-7.47; 9.34]
8.97 [-1.61; 19.55]
2.82 [-4.94; 10.57]
1.88 [-7.18; 10.94]
9.95 [-4.36; 24.25]
3.79 [-7.41; 15.00]
2.86 [-10.01; 15.73]
0.98 [-10.38; 12.34]
10.61 [0.11; 21.11]
4.45 [-2.37; 11.28]
3.52 [-5.23; 12.26]
1.64 [-6.00; 9.27]
0.66 [-10.44; 11.76]
A summary of the results, considering the quality of the evidence and evaluation periods, for each outcome is shown in Table 4. For the other scores (SST, WORC, and SPADI), no significant effects were observed in any comparison or period.
Safety of sub-acromial injections
Ten studies in this systematic review evaluated the occurrence of adverse effects. Six of these, by Alvarez et al[34], Lin et al[36], Chang et al[37], Lin et al[42], Meloni et al[43], and Cai et al[44], did not report any adverse effects with the sub-acromial injection of PRP, HA, corticosteroids, and/or prolotherapy. The most observed adverse effect was local pain, but none of the events required hospitalisation. The scarcity of data precluded any between-group comparisons.
Assessment of tendon morphology with complementary imaging examinations
Lin et al[36], Chang et al[37], Cole et al[39], Lin et al[42], Meloni et al[43], Cai et al[44], Shams et al[48], Schwitzguebel et al[49], Kwong et al[50], and Dadgostar et al[51] evaluated tendon morphology using imaging examinations. Only four of these studies observed significant differences between the groups undergoing sub-acromial infiltrations throughout the follow-up period. Due to the absence of standardised metrics for morphological assessment in most studies, quantitative assessments regarding this outcome could not be established.
DISCUSSION
The interest in the application of regenerative medicine for the treatment of orthopaedic injuries, including osteoarthritis, chondropathies, and ligament and tendon injuries, has increased in the last decade[58,59]. Injections are considered minimally invasive, safe, and accessible. Although many studies have been conducted on rotator cuff injuries, the benefits of different injectables in this context remain unclear.
Zhi et al[19] included studies published in the last decade, totalling 10 RCTs (861 patients) that compared the following injection (sub-acromial and intra-articular techniques) options: Prolotherapy, HA, PRP, and corticosteroids. No studies that compared these medicines and biological products with a placebo were selected, which is a major limitation, given the unclear superiority of any of the agents over placebo.
This review showed that in the short term, HA demonstrated the most favourable outcomes; however, in the medium and long terms, HA, PRP, the combination of HA + PRP, and corticosteroids did not show superior efficacy compared to placebo in terms of pain reduction. Regarding functional improvement, the combination of HA + PRP proved to be more effective than placebo over the medium and long terms.
These findings do not converge with those of our systematic review, possibly because of two main factors: Differences in RCT selection criteria and non-fixed evaluation periods. Many studies have already demonstrated the effectiveness of corticosteroids for rotator cuff injuries and other shoulder diseases, such as adhesive capsulitis, but with effects that are not sustainable in the medium and long terms[60-62].
Lin et al[63] published a systematic review of 18 RCTs (1495 patients), including patients diagnosed with rotator cuff injury, without the need for confirmation by imaging examination, and considered short-, medium-, and long-term follow-up periods as 3-6 weeks, 12 weeks, and > 24 weeks, respectively. The injection options included corticosteroids, HA, PRP, non-steroidal anti-inflammatory drugs (NSAIDs), botulinum toxin, and prolotherapy. Regarding pain improvement, the corticosteroids were superior to the placebo (clinically insignificant) only in the short term. In network meta-analyses, the difference between treatments (HA, NSAID, PRP, and corticosteroids) and placebo was not significant in the short or medium term. However, in the long term, prolotherapy appeared to be the best treatment according to the probability of classification (100%). The post-intervention function and meta-analyses showed that corticosteroids were superior to placebo only in the short term, whereas PRP was superior to placebo in the long term. This discrepancy in findings between the present and previous reviews may be due to the differences in study eligibility criteria and the period of outcome evaluation. Herein, HA was compared with prolotherapy only through an indirect analysis. Regarding pain, HA was superior to prolotherapy in the short term, but data for comparisons in the medium and long terms within the applied criteria were not available. Regarding the short- and medium-term functional outcomes, prolotherapy was not superior to placebo.
Although different methodologies were used, other reviews such as those by Chang et al[64] and Barman et al[65] showed the effectiveness of HA and PRP, respectively, during the evaluation period. Jiang et al[66] investigated the use of HA, PRP, and corticosteroid injections in traumatic and degenerative rotator cuff injuries, including full-thickness tears, and observed that HA plays a role in short-term functional and pain improvement, with PRP being more effective after 6 months of follow-up. This study did not assess the effects of combining HA with PRP in the meta-analyses.
In our systematic review, no significant improvement in pain was observed, based on the VAS analysis, after 3 months. The International Association for the Study of Pain defines pain as 'an unpleasant sensory and emotional experience associated with, or resembling that associated with, actual or potential tissue damage[67]. Therefore, pain is inherently subjective and potentially multifactorial. The divergent results of sustained functional improvement vs the perception of pain improvement in the short term may be due to the functional scores which consider, in addition to pain, the objective parameters such as range of motion and strength. Thus, in clinical practice, for evaluating outcomes and planning treatment, considering functional scores may be more relevant in conjunction with the VAS.
Regarding the complications due to the injections, the data from the primary studies indicate that the medications and biologics evaluated were safe, with a reported low occurrence of side effects. Regarding the economic analysis of injections, few studies have specifically assessed the topic in relation to rotator cuff tendinopathies. Ranawat et al[68] evaluated the costs involved in the treatment of knee osteoarthritis and demonstrated that HA offers better cost-effectiveness compared to alternative treatments (such as physiotherapy or medication) and delays the need for arthroplasty. In another analysis, comparing PRP and HA in the treatment of knee osteoarthritis, PRP injections were more economically advantageous than those with HA[69]. Eubank et al[70] compared the main treatment approaches for partial rotator cuff tears, including PRP, HA, corticosteroid, prolotherapy, and anaesthetic injections, and concluded the absence of superiority of surgical treatment over non-surgical treatment, with surgery being a more invasive, costly option with a higher risk of complications.
This systematic review captured many studies on the applicability of injections for rotator cuff injuries. However, many of these publications lacked adequate methodology and often encompassed many shoulder diseases with different pathophysiologies, such as rotator cuff injuries, adhesive capsulitis, and shoulder osteoarthritis, all of which were treated with the same modality. Added to this problem is the fact that many clinical trials that investigate this topic include participants with only a clinical diagnosis of a rotator cuff injury, introducing a selection bias that favours more severe injuries (such as total tendon tears) or other concomitant conditions.
Consequently, the need for higher-quality RCTs remains. These studies should include patients with imaging examinations that confirm the diagnosis of rotator cuff injury and those who undergo US-guided sub-acromial injections for maximum effectiveness. Furthermore, given the diversity of treatment options available, investigators should report agent-specific factors, including the molecular weight of HA or the composition of PRP, to help ensure accurate comparisons. In addition, follow-up duration should be carefully considered, given the different outcomes observed in the short, medium, and long terms. Further, follow-up after sub-acromial injections must include imaging evaluations using standardised metrics to elucidate the morphological changes caused by these medications and biological products.
Finally, considering the data found in this systematic review, future research that includes the study of the adjuvant actions of substances may yield better and more robust results. Although the synergistic effects remain unclear, the combined use of substances has the potential to produce long-lasting improvements in pain and function. Currently, some studies show that these adjuvant actions of substances can lead to better outcomes, which are also sustained in the medium and long terms. For example, in the study by Cai et al[44], the use of HA in combination with PRP was shown to be more effective both in pain and functional scores.
CONCLUSION
According to the results obtained from this systematic review, subacromial injections of HA and HA + PRP for rotator cuff injuries may assist in short-term pain control and functional improvement in the medium and long terms, demonstrating superiority over the other options analysed.
Footnotes
Provenance and peer review: Unsolicited article; Externally peer reviewed.
Peer-review model: Single blind
Corresponding Author's Membership in Professional Societies: Brazilian Society of Orthopedics and Traumatology, No. 17433.
Specialty type: Orthopedics
Country of origin: Brazil
Peer-review report’s classification
Scientific Quality: Grade B
Novelty: Grade B
Creativity or Innovation: Grade B
Scientific Significance: Grade B
P-Reviewer: Zhang GH S-Editor: Qu XL L-Editor: A P-Editor: Wang WB
Zhi F, Cai F, Zhang W, Xiong L, Hu J, Lin X. Clinical efficacy of different shoulder joint drug injections for rotator cuff injuries: A network meta-analysis.Medicine (Baltimore). 2022;101:e30659.
[PubMed] [DOI][Cited in This Article: ][Reference Citation Analysis (0)]
Page MJ, McKenzie JE, Bossuyt PM, Boutron I, Hoffmann TC, Mulrow CD, Shamseer L, Tetzlaff JM, Akl EA, Brennan SE, Chou R, Glanville J, Grimshaw JM, Hróbjartsson A, Lalu MM, Li T, Loder EW, Mayo-Wilson E, McDonald S, McGuinness LA, Stewart LA, Thomas J, Tricco AC, Welch VA, Whiting P, Moher D. The PRISMA 2020 statement: an updated guideline for reporting systematic reviews.BMJ. 2021;372:n71.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 32381][Cited by in RCA: 32603][Article Influence: 8150.8][Reference Citation Analysis (1)]
Hudak PL, Amadio PC, Bombardier C. Development of an upper extremity outcome measure: the DASH (disabilities of the arm, shoulder and hand) [corrected]. The Upper Extremity Collaborative Group (UECG).Am J Ind Med. 1996;29:602-608.
[PubMed] [DOI][Cited in This Article: ][Cited by in RCA: 37][Reference Citation Analysis (0)]
Chandler J, Cumpston M, Thomas J, Higgins JPT, Deeks JJ, Clarke MJ.
Chapter I: Introduction. In: Higgins JPT, Thomas J, Chandler J, Cumpston M, Li T, Page MJ, Welch VA (editors). Cochrane Handbook for Systematic Reviews of Interventions version 6.5, 2024 [cited 2025 Jan 9]. Available from: https://training.cochrane.org/handbook/current/chapter-08.
[PubMed] [DOI][Cited in This Article: ]
Guyatt GH, Oxman AD, Vist GE, Kunz R, Falck-Ytter Y, Alonso-Coello P, Schünemann HJ; GRADE Working Group. GRADE: an emerging consensus on rating quality of evidence and strength of recommendations.BMJ. 2008;336:924-926.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 11058][Cited by in RCA: 13980][Article Influence: 822.4][Reference Citation Analysis (0)]
Alvarez CM, Litchfield R, Jackowski D, Griffin S, Kirkley A. A prospective, double-blind, randomized clinical trial comparing subacromial injection of betamethasone and xylocaine to xylocaine alone in chronic rotator cuff tendinosis.Am J Sports Med. 2005;33:255-262.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 115][Cited by in RCA: 105][Article Influence: 5.3][Reference Citation Analysis (0)]
Hsieh LF, Lin YJ, Hsu WC, Kuo YC, Liu YC, Chiang YP, Wang CP. Comparison of the corticosteroid injection and hyaluronate in the treatment of chronic subacromial bursitis: A randomized controlled trial.Clin Rehabil. 2021;35:1305-1316.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 3][Cited by in RCA: 4][Article Influence: 1.0][Reference Citation Analysis (0)]
Damjanov N, Barac B, Colic J, Stevanovic V, Zekovic A, Tulic G. The efficacy and safety of autologous conditioned serum (ACS) injections compared with betamethasone and placebo injections in the treatment of chronic shoulder joint pain due to supraspinatus tendinopathy: a prospective, randomized, double-blind, controlled study.Med Ultrason. 2018;20:335-341.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 21][Cited by in RCA: 24][Article Influence: 3.4][Reference Citation Analysis (0)]
Meloni F, Milia F, Cavazzuti M, Doria C, Lisai P, Profili S, Meloni GB. Clinical evaluation of sodium hyaluronate in the treatment of patients with sopraspinatus tendinosis under echographic guide: experimental study of periarticular injections.Eur J Radiol. 2008;68:170-173.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 32][Cited by in RCA: 39][Article Influence: 2.2][Reference Citation Analysis (0)]
Schwitzguebel AJ, Kolo FC, Tirefort J, Kourhani A, Nowak A, Gremeaux V, Saffarini M, Lädermann A. Efficacy of Platelet-Rich Plasma for the Treatment of Interstitial Supraspinatus Tears: A Double-Blinded, Randomized Controlled Trial.Am J Sports Med. 2019;47:1885-1892.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 41][Cited by in RCA: 51][Article Influence: 8.5][Reference Citation Analysis (0)]
Kwong CA, Woodmass JM, Gusnowski EM, Bois AJ, Leblanc J, More KD, Lo IKY. Platelet-Rich Plasma in Patients With Partial-Thickness Rotator Cuff Tears or Tendinopathy Leads to Significantly Improved Short-Term Pain Relief and Function Compared With Corticosteroid Injection: A Double-Blind Randomized Controlled Trial.Arthroscopy. 2021;37:510-517.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 28][Cited by in RCA: 66][Article Influence: 16.5][Reference Citation Analysis (0)]
Tashjian RZ, Deloach J, Porucznik CA, Powell AP. Minimal clinically important differences (MCID) and patient acceptable symptomatic state (PASS) for visual analog scales (VAS) measuring pain in patients treated for rotator cuff disease.J Shoulder Elbow Surg. 2009;18:927-932.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 386][Cited by in RCA: 488][Article Influence: 30.5][Reference Citation Analysis (0)]
Kanto K, Lähdeoja T, Paavola M, Aronen P, Järvinen TLN, Jokihaara J, Ardern CL, Karjalainen TV, Taimela S. Minimal important difference and patient acceptable symptom state for pain, Constant-Murley score and Simple Shoulder Test in patients with subacromial pain syndrome.BMC Med Res Methodol. 2021;21:45.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 5][Cited by in RCA: 12][Article Influence: 3.0][Reference Citation Analysis (0)]
Chang Y, Philips S, Scholl E, Bains S, Bhandari M.
Hyaluronic acid (HA) injection versus control for painful shoulder conditions: A systematic review and meta-analysis of randomized controlled trials. OE Original 2022; 5: 2. Available from: https://myorthoevidence.com/Blog/Show/199.
[PubMed] [DOI][Cited in This Article: ]
Barman A, Mishra A, Maiti R, Sahoo J, Thakur KB, Sasidharan SK. Can platelet-rich plasma injections provide better pain relief and functional outcomes in persons with common shoulder diseases: a meta-analysis of randomized controlled trials.Clin Shoulder Elb. 2022;25:73-89.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 1][Cited by in RCA: 1][Article Influence: 0.3][Reference Citation Analysis (0)]
Jiang X, Zhang H, Wu Q, Chen Y, Jiang T. Comparison of three common shoulder injections for rotator cuff tears: a systematic review and network meta-analysis.J Orthop Surg Res. 2023;18:272.
[PubMed] [DOI][Cited in This Article: ][Reference Citation Analysis (0)]
Ranawat A, Guo K, Phillips M, Guo A, Niazi F, Bhandari M, Waterman B. Health Economic Assessments of Hyaluronic Acid Treatments for Knee Osteoarthritis: A Systematic Review.Adv Ther. 2024;41:65-81.
[PubMed] [DOI][Cited in This Article: ][Reference Citation Analysis (0)]
Bendich I, Rubenstein WJ, Cole BJ, Ma CB, Feeley BT, Lansdown DA. What Is the Appropriate Price for Platelet-Rich Plasma Injections for Knee Osteoarthritis? A Cost-Effectiveness Analysis Based on Evidence From Level I Randomized Controlled Trials.Arthroscopy. 2020;36:1983-1991.e1.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 13][Cited by in RCA: 16][Article Influence: 3.2][Reference Citation Analysis (0)]
Eubank BHF, Sheps DM, Dennett L, Connick A, Bouliane M, Panu A, Harding G, Beaupre LA. A scoping review and best evidence synthesis for treatment of partial-thickness rotator cuff tears.J Shoulder Elbow Surg. 2024;33:e126-e152.
[PubMed] [DOI][Cited in This Article: ][Cited by in Crossref: 2][Reference Citation Analysis (0)]
