The risks and benefits of glucocorticoid treatment for tendinopathy: A systematic review of the effects of local glucocorticoid on tendon

Seminars in Arthritis and Rheumatism ] (2013) ]]]–]]]

Contents lists available at ScienceDirect

Seminars in Arthritis and Rheumatism journal homepage: www.elsevier.com/locate/semarthrit

The risks and benefits of glucocorticoid treatment for tendinopathy: A systematic review of the effects of local glucocorticoid on tendon☆ Benjamin John Floyd Dean, MRCSn, Emilie Lostis, BSc, Thomas Oakley, BM, BSc, Ines Rombach, MSc, Mark E. Morrey, MD, Andrew J. Carr, FRCS Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences (NDORMS), Botnar Research Centre, Institute of Musculoskeletal Sciences, Nuffield Orthopaedic Centre, Windmill Rd, Oxford OX3 7LD, UK

a r t i c l e in fo

Keywords: Glucocorticoid Tendon Tenocyte Steroid Fibroblast

a b s t r a c t Objective: Our primary objective was to summarise the known effects of locally administered glucocorticoid on tendon tissue and tendon cells. Methods: We conducted a systematic review of the scientific literature using the PRISMA and Cochrane guidelines of the Medline database using specific search criteria. The search yielded 50 articles, which consisted of 13 human studies, 36 animal studies and one combined human/animal study. Results: Histologically, there was a loss of collagen organisation (6 studies) and an increase in collagen necrosis (3 studies). The proliferation (8 studies) and viability (9 studies) of fibroblasts was reduced. Collagen synthesis was decreased in 17 studies. An increased inflammatory cell infiltrate was shown in 4 studies. Increased cellular toxicity was demonstrated by 3 studies. The mechanical properties of tendon were investigated by 18 studies. Descriptively, 6 of these studies showed a decrease in mechanical properties, 3 showed an increase, while the remaining 9 showed no significant change. A meta-analysis of the mechanical data revealed a significant deterioration in mechanical properties, with an overall effect size of
0.67 (95% CI ¼ 0.01 to
1.33) (data from 9 studies). Conclusions: Overall it is clear that the local administration of glucocorticoid has significant negative effects on tendon cells in vitro, including reduced cell viability, cell proliferation and collagen synthesis. There is increased collagen disorganisation and necrosis as shown by in vivo studies. The mechanical properties of tendon are also significantly reduced. This review supports the emerging clinical evidence that shows significant long-term harms to tendon tissue and cells associated with glucocorticoid injections. & 2013 Elsevier Inc. All rights reserved.

Introduction In September 1948 at the Mayo clinic, cortisone was injected into a patient for the first time in the treatment of rheumatoid arthritis to dramatic effect [1]. The 1950 Nobel Prize in Physiology or Medicine was awarded jointly to Edward Calvin Kendall, Tadeus Reichstein and Philip Showalter Hench directly relating to this work “for their discoveries relating to the hormones of the adrenal

☆ The authors of this work are funded by the Musculoskeletal Biomedical Research Unit of the National Institute for Health Research (B.D., M.M., E.L., T.O. and A.C.), the Jean Shanks Foundation (B.D.) and Orthopaedic Research UK (B.D.). The funding sources had no role in the study design, collection, analysis and interpretation of data; in the writing of the article; and in the decision to submit the manuscript for publication. n Corresponding author at: Nuffield Department of Orthopaedics, Rheumatology and Musculoskeletal Sciences (NDORMS), Botnar Research Centre, Institute of Musculoskeletal Sciences, Nuffield Orthopaedic Centre, Windmill Rd, Oxford OX3 7LD, UK. E-mail address: [email protected] (B.J.F. Dean).

0049-0172/$ - see front matter & 2013 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.semarthrit.2013.08.006

cortex, their structure and biological effects.” The use of glucocorticoid in the treatment of painful musculoskeletal disease has since proliferated to the point that now in the UK over 500,000 intraarticular glucocorticoid injections (GCIs) are administered per year in the primary care setting [2]. GCIs are used to relieve pain and/or inflammation in a wide variety of musculoskeletal disorders including osteoarthritis, inflammatory arthritis, tenosynovitis, tendinopathy and degenerative spine disease. The evidence regarding the clinical efficacy of GCIs is conflicting but broadly shows some short-term benefits in terms of pain relief [3–6]. For example, in the treatment of shoulder pain, trials have shown only short-term benefits with no significant long-term gains [6–8]. Emerging highquality evidence also points to poorer long-term outcomes associated with GCIs in the treatment of tendinopathy [9]. GCIs are frequently applied in close proximity to tendons with common examples including the rotator cuff, the flexor and extensor tendon origins around the elbow, the gluteus medius, the Achilles and the patellar tendons, the flexor tendons in the hand (i.e., trigger finger) and the extensor tendons around the

B.J.F. Dean et al. / Seminars in Arthritis and Rheumatism ] (2013) ]]]–]]]

This systematic review used the PRISMA-Statement (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) and the Cochrane handbook as guidelines in the development of the study protocol and the report of the current study [25,26]. The inclusion criteria and methods of analysis were specified in advance and documented in a protocol. Studies were identified using the Medline electronic database. No limit was placed on the year of data entry, but in practice, there were no results prior to 1956. The search was undertaken in June 2013. The following search terms were used: steroid OR corticosteroid OR glucocorticoid AND tendon OR rotator cuff OR Achilles OR tendinn OR tenocyte. Additional studies were located by searching papers referenced in listed articles. The studies identified by the searches were combined, and duplicates were excluded. The abstracts were initially screened before analysis of the selected full-text articles. Full inclusion/exclusion criteria are detailed in Appendix 1. Studies had to relate to the use of local glucocorticoid on tendon tissue or tendon cells. Review articles and case studies were excluded. Papers pertaining to steroids other than glucocorticoids, such as anabolic steroids, were excluded. Those articles addressing steroid use other than for a peri-tendinous or tendinous injection, such as intra-articular steroid injection, were excluded. Studies using systemic steroid as opposed to injected corticosteroid were excluded. Any study without results relating to histological, cellular, molecular or mechanical tissue changes was excluded. If a study could not be obtained in English, it was excluded.

The search strategy yielded 4424 results (Fig. 1). After the exclusion of duplicates and review articles, there were 1996 articles. Screening the articles revealed 40 articles on humans and 38 articles on animals that met the criteria based on their abstracts. Further assessment of eligibility, based on full-text articles, led to the exclusion of 28 of these 78 papers. The reasons for the exclusion of these 28 papers were as follows: no control group [6], reviews [12], systemic glucocorticoid therapy [6] and not related to tendon [4]. This left 50 articles meeting our inclusion criteria, and they are summarised in Appendix 3. Study characteristics Of the 50 included articles, 36 related to animal studies, 13 to human studies and 1 to a study that was on both animals and humans. Of the 36 animal studies, 25 were in vivo and 11 in vitro; while of the 13 human studies, 12 were in vitro, 1 was in vivo, and 1 study was both in vivo and in vitro. The 1 combined human and animal study was in vitro. The most common animal used was rat (19 studies), followed by rabbit (9 studies), chick embryo (5 studies), dog (2 studies), cow (1 study) and multiple animals (1 study). The 25 in vivo animal studies used Achilles tendon

Flow chart of systemac review protocol Idenficaon

Search strategies

Study selection

Screening

Methods

The search, selection of studies and data analysis were performed independently by 2 individuals (T.O. and BD for the articles on humans and E.L. and B.D. for the articles on animals). Agreement on inclusion was achieved after review of the full-text articles and a joint decision by both individuals based on the inclusion/exclusion criteria. The data were then extracted using a spreadsheet designed by 2 authors (B.D. and E.L.), this included data relating to study heterogeneity and methodological quality. The data extracted included study subject characteristics, glucocorticoid used, source of cultured cells and cells used method of tissue analysis, control group, results and statistical methods. Methodological quality was assessed using an 8-point scoring system (Appendix 2) based on the method used by Hegedus et al. [27].

Analysis

wrist (i.e., De Quervain's tenosynovitis). It has been recurrently postulated that there is an increased risk of tendon rupture associated with GCI [10] but no high-quality evidence exists to adequately confirm or refute this hypothesis [11,12]. It is important to remember that GCI is often used in the context of an abnormal diseased tendon in which the risk of rupture is already increased. However, there is strong evidence that oral corticosteroids are associated with a higher risk of tendon rupture [13], and an increased spinal fracture risk associated with epidural GCIs has also recently been reported [14]. The mechanisms of action of glucocorticoids are multiple, highly complex and incompletely understood [15,16]. One important pathway involves the activation of specific cytoplasmic glucocorticoid receptors, which then migrate to the cell nucleus to affect gene transcription. Generally, glucocorticoids are thought to be anti-inflammatory, but the reality may not be so simple [17]. The tendon changes that occur in painful human tendinopathy are generally considered to be consistent with a failed healing response [18,19]. Normal tendon healing occurs with sequential inflammatory, proliferative and remodelling phases [20]. Fibroblast proliferation, angiogenesis and nerve ingrowth are all important in the healing process [21,22]. Tendinopathy is characterised by abnormal tenocyte morphology and disorganised collagen architecture [19]. Although the presence of inflammation in tendinopathy has been proposed by some authors [23], few studies have shown the presence of a “classical” inflammatory process involving the inward migration of inflammatory cells driven by inflammatory mediators [24]. Therefore, the logic of using GCIs in the treatment of a tendinopathy is not convincing. In this context, the purpose of this review was to determine the effects of local GCI on both tendon tissue and tendon cells. We aimed to describe and summarise the histological, molecular and mechanical changes.

Included

2

Papers idenfied through database searching (n=4424) Aer removal of duplicates and review papers (n=1996)

Screening of paper abstracts

Human papers pre analysis (n=40)

Animal papers pre analysis (n=38)

Review of full text and applicaon of exclusion criteria

Human papers included (n=13)

Combined animal and human papers included (n=1)

Animal papers included (n=36)

Fig. 1. Flow chart of systematic review protocol.

B.J.F. Dean et al. / Seminars in Arthritis and Rheumatism ] (2013) ]]]–]]]

(11 studies), shoulder tendon (5 studies), forearm tendon (3 studies), peroneal tendon (2 studies), tail tendon (2 studies), patellar tendon (1 study) and both Achilles and patellar tendon (1 study). The human in vivo study used Achilles tendon, while the combined human in vivo/vitro study used shoulder tendon for its in vivo component. Of the 14 human studies, the glucocorticoid/s used were dexamethasone alone (6 studies), triamcinolone (3 studies), dexamethasone and triamcinolone (2 studies), dexamethasone and methylprednisolone (1 study) and hydrocortisone (1 study). Of the 36 animal studies, the glucocorticoid/s used were hydrocortisone (9 studies), dexamethasone (8 studies), methylprednisolone (7 studies), betamethasone (6 studies), triamcinolone (4 studies), methylprednisolone and betamethasone (1 study) and prednisolone (1 study). The animal and human study used triamcinolone.

Table 1 Histological changes Human studies General cell Viability ↓ [44–51,90] characteristics Proliferation ↓ [45,48–50,52,53] Collagen Collagen necrosis ↑a [56] Collagen organisation ↓a morphology [56] Cellular Inflammatory cells ↑a [56] Inflammatory cells ↑[57,59] migration ↑a[58] Fibroblast migration ↓ [52] Cellular toxicity Adhesions and others

Study methodology and assessing the risk of bias All included studies stated described their study group and control groups and clearly described method/s of tissue analysis. The results of the methodological quality assessment are detailed in Appendix 3. The median score of the human studies was 8 (range 5–8). The median score of the animal studies was 6 (range 4–8). Of the 50 studies, 40 produced quantitative and/or semiquantitative results with a stated statistical significant at p o 0.05. Of the 10 studies that did not state statistical significance, 3 were descriptive histological studies (1 human in vivo and 2 animal in vivo), 2 were quantitative mechanical and descriptive histological studies (both animal in vivo) and the remaining 5 used solely quantitative assays (all animal in vitro with 1 also involving human in vitro). Studies that did not have control groups for comparison were excluded. Studies that used descriptive non-quantitative methods only and/or studies that did not state the statistical significance of their results were included in results; these results were clearly marked with “a”. This methodological assessment means that the results have not included studies with a large degree of bias, and that those with higher degrees of potential bias have been highlighted to readers. The study heterogeneity precluded a meta-analysis of the histological and molecular results. However the studies relating to mechanical properties were systematically analysed to obtain data sets for a meta-analysis. Nine of the 17 studies provided adequate data from which the effect size with 95% upper and lower confidence intervals could be calculated. The data set extracted from each study was based upon which property of mechanical strength (i.e., yield stress, yield energy and modulus of stiffness) was quoted as the study's main finding in its abstract. More than 1 data set was included per study if each data set related to a distinct experimental subgroup.

3

a

a

↑ [47,48] ↑ [49] Altered collagen birefringencea [56] Adipocyte clustering around steroid [52] Non tenocyte differentiation of hTSCs [53]

Animal studies Proliferation ↓ [54,55]

Collagen necrosis ↑ [57] ↑a[58] Collagen organisation ↓[29,57,59,60] ↓a [40]→[89]

Fibroblast migration ↑ [60] ↓ [61] ↑a [49]→[88] Adhesions ↑ [62,63] ↓ [64]

Denotes that study did not state statistical significance of result.

increase in collagen necrosis (3 studies). The proliferation (8 studies) and viability (9 studies) of fibroblasts was reduced. An increased inflammatory cell infiltrate was shown in 4 studies. Increased cellular toxicity was demonstrated by 3 studies. Results regarding fibroblast migration and adhesions were conflicting. Collagen synthesis was decreased in 17 studies. An increased ratio of type 3 to type 1 collagen was shown in 2 studies. Apoptosis was unaffected in 3 studies and increased in 3 studies. Small numbers of studies had demonstrated changes in matrix enzymes (MMPs/ TIMPs), proteoglycans, cytokines and other substances including FOX-01 and Sirtuin-1. Eighteen studies investigated the mechanical properties of tendon (Table 3). Descriptively, 6 of these studies showed a decrease in mechanical properties, 3 showed an increase, while the remaining 9 showed no significant change. Nine studies provided adequate in vivo data from which to calculate the effect size with upper and lower 95% confidence intervals. Three studies contributed more than 1 data set as they had obtained results for more than 1 group of GCI-treated animals; Mikolyzk et al. [29] analysed 3 groups of animal at different time points; Oxlund et al. [30] analysed the effects on 2 different tendons and Plotkin et al. [31] analysed the effects of 2 different GCI dose. Figure 2

Table 2 Molecular changes Human studies

Animal studies

↓ [45,47–50,52,53,65]

↓ [55,66–68] ↓a [69–73] ↑ [74,75]

Statistics Collagen synthesis

All statistics and the forest plot [28] were carried out using Microsoft Excel 2007TM (Microsoft Corporation, Redmond, WA). Where appropriate, data were combined using meta-analytical methods. Meta-analyses and Forest Plots were performed in Stata version 12 (Stata, College Station, TX). Random effects metaanalysis was performed to account for the heterogeneity between studies.

Collagen ratio type III–I Apoptosis MMPs/TIMPs

Cytokines and others

Results There were significant histological and molecular changes after local glucocorticoid administration (Tables 1 and 2). Histologically, there was a loss of collagen organisation (6 studies) and an

a

↑ [47,51,90] - [45–47] ↑ TIMP [52] ↓MMP 2/8/9/13 [52] ↓β-Integrin production [47] ↑ MMP 1/13 [47] ↑ SDF-1 [65] ↑ ROS [48] ↑ FOX-01 [48] ↓ Proteoglycan [76] ↑ PPAR-γ and SOX-9 [53] ↑ Senescence [46,77] ↓ Sirtuin-1 [77]

MMP2/9 - [75] TIMP1/2 - [75]

↑ Aggrecan [75] ↑ Fibronectin [75] ↓ TNFα [75]

Denotes that study did not state statistical significance of result.

B.J.F. Dean et al. / Seminars in Arthritis and Rheumatism ] (2013) ]]]–]]]

4

Table 3 Mechanical changes References

Descriptive changes

Overall effect on mechanical properties

Martins et al. [78] Haraldsson et al. [79,87]

No difference in maximum failure load or absorbed energy Reduction in tensile fascicle yield strength and Young's modulus; unaffected strain properties, peak stress and fascicle diameter Reduced mechanical properties in steroid groups at 1 week Increased energy and load to failure, but no difference in material stiffness or strain Decreased failure stress and total energy absorbed; increased total strain No difference in elastic modulus, ultimate load and ultimate stress No difference in mean separation forces Reduced failure load and energy to failure, unchanged strain to failure No difference in yield load and yield stress Increased maximum stress for peroneus brevis; increased elastic stiffness and maximum load for peroneus longus Initial reduction in failure load; no difference in failure load after 2 weeks No difference in yield load, relative yield load and stiffness No difference in yield load, stiffness or strain No difference in failure load, stiffness or failure site Decreased modulus of elastic stiffness Increased tensile strength No difference in tensile strength

↓

Mikolyzk et al. [29] Shapiro et al. [62] Hugate et al. [80] Martin et al. [81] McWhorter et al. [82] Kapetanos [64] Oxlund [66] Oxlund [30] Kennedy and Willis [40] Plotkin et al. [31] Mackie et al. [83] Matthews et al. [84] Unverferth and Olix [60] Ketchum [85] Gonzalez [86] a

↓ ↑ ↓ ↓ ↑ ↓a ↓ ↑ -a

Denotes that study did not state statistical significance of result.

represents this forest plot. The overall effect size was
0.67 (95% confidence interval
0.01 to
1.33, p ¼ 0.046), demonstrating that in these 9 studies there was a clear trend towards reduced mechanical properties in tendon after glucocorticoid injection.

Discussion Overall the results are broadly negative, both in terms of the effects on several specific cellular characteristics and on the mechanical properties of tendon. While some studies have shown short-term pain relief, these results provide plausible mechanisms by which glucocorticoid treatment may result in adverse patient outcomes in the treatment of degenerative tendinopathy particularly in the long term [9]. The cause of tendinopathy has been the subject of much heated debate over the years, [32] and many different theories have been

postulated [33–36]. The histological and molecular changes in tendinopathy are undoubtedly consistent with mechanical tendon failure and a persistent failed healing response [18,19,37]. The inflammatory element of tendinopathy appears more consistent with a failing attempt to heal than a classical “inflammatory response” involving an inward migration of inflammatory cells. This apparent attempt at healing involving an inflammatory component appears to be more likely present in early tendinopathy [38] and decreasingly present as disease progresses, as tendon becomes progressively hypoxic and abnormal [39]. In this context, the consistent negative findings summarised in this review, both in terms of the changes to the cellular and tendon properties, should be of great cause for concern. Overall glucocorticoid appears to have a negative effect on tendon homoeostasis with increased collagen disorganisation and collagen necrosis seen after treatment. Indeed it may be hypothesised that the increased number of inflammatory cells present may

Author

Measure

SMD (95% CI)

Haraldsson 2009

yield stress

-2.38 (-3.44, -1.31)

Kapetanos 1982

yield load

-1.41 (-2.40, -0.42)

Martin 1999

yield stress

-0.46 (-1.39, 0.48)

Martins 2011

yield load

-1.00 (-1.76, -0.24)

Matthews 1974

yield load

-0.39 (-1.01, 0.23)

Mikolyzk 2009

yield load (timepoint 1)

-12.73 (-16.76, -8.70)

Mikolyzk 2009

yield load (timepoint 3)

0.00 (-0.86, 0.86)

Mikolyzk 2009

yield load (timepoint 5)

-1.67 (-2.70, -0.63)

Oxlund 1980

yield load (peroneus brevis)

3.52 (1.82, 5.22)

Oxlund 1980

yield load (peroneus longus)

1.76 (0.45, 3.08)

Plotkin 1976

yield load (high dose)

-0.17 (-0.58, 0.24)

Plotkin 1976

yield load (low dose)

-0.44 (-0.86, -0.02)

Unverfeth 1973

modulus of stiffness

-1.18 (-2.55, 0.19)

Overall (I-squared = 87.3%, p = 0.000)

-0.67 (-1.33, -0.01)

NOTE: Weights are from random effects analysis -16.8 Mechanical properties reduced

16.8

0 Mechanical properties increased

Fig. 2. Forest plot of studies analysing effects of glucocorticoid on the mechanical properties of tendon (X axis—effect size (standardised mean difference) and Y axis—studies included).

B.J.F. Dean et al. / Seminars in Arthritis and Rheumatism ] (2013) ]]]–]]]

be the result of a reparative response to the glucocorticoid-induced tendon damage. This explanation is consistent with the result of the mechanical studies that show a short-term deterioration in the mechanical properties [29,40] which then recovers in the longer term as a result of the healing response to the glucocorticoidinduced damage. The mechanisms for such changes in tendon are unclear with the classical description of glucocorticoid as an “antiinflammatory” agent appearing to be an over simplification [17]. Glucocorticoid-induced senescence is one mechanism by which long-term degenerative changes in tendon tissue may be worsened. The glucocorticoid reduction in collagen synthesis is another important mechanism by which altered homoeostasis may lead to deterioration in the mechanical properties of tendon, thus potentially increasing levels of future degeneration. It is worth noting that there is still a clear role for GCI in the treatment of tendinopathy, but that it is vital to consider the potential negative effects highlighted by this review when making one's clinical decision on a case-by-case basis. Given the huge variability in terms of both patient characteristics and the pathogeneses of the different tendinopathies, it is beyond the scope of this review to give prescriptive management advice to clinicians regarding the use of GCIs. Certainly the repeated use of GCIs in younger patients with an “overuse” type of tendinopathy appears unwise and potentially harmful, given the negative effects of glucocorticoid on the tendon healing. However, the use of GCIs for short-term clinical gains in older patients with degenerate tendons who are not suitable surgical candidates still appears a very appropriate treatment strategy. While the judicious use of GCIs for specific conditions such as trigger finger and De Quervain's tenosynovitis, in which the anti-proliferative and antiinflammatory effects are of definite therapeutic benefit, is still a very effective and justifiable treatment strategy [41,42]. The use of radiologically guided injections is becoming more commonplace and evidence suggests that they are significantly more accurate than “blind” injections [43]. Whether this increased accuracy translates into reduced complication rates and better clinical outcomes is something that needs to be determined by future research. Limitations of this review The results and generalisable meaning of this review are both limited by the quality and heterogeneity of the included studies. There was significant study heterogeneity in terms of study type (animal/human or in vivo/vitro), the anatomical source of tendon/ tendon cells, glucocorticoid used, dose of glucocorticoid, mode of tissue analysis, time points and control type. It is particularly important to appreciate that several assumptions have been made in carrying out the meta-analysis of the mechanical studies. One had to assume that the extracted data was accurate (means and standard deviations) and that their underlying distributions were normal. As a result, some caution should be exercised when interpreting the results of the meta-analysis. For the purposes of this review, we have asked a specific but general research question and included only studies that met criteria specific to this. However, our objective was to synthesise the overall research findings in this broad area and a degree of study heterogeneity had to be accepted in order to achieve this. The degree to which this review's results are generalisable to human patients is certainly open to debate. In vitro findings often conflict with in vivo findings; however, our review's key findings appear broadly consistent between the different study types. The degree to which the results of the studies were descriptive, semi-quantitative or quantitative was highly variable. As a result, it has been made clear when results have not been shown to be statistically significant (Tables 1–3). The variable blinding of the

5

observers undertaking the semi-quantitative tissue grading does increase the risk of study bias towards positive findings. The measurement of multiple mechanical characteristics of tendon, and the absence of defining a primary study outcome, combined to result in some likely bias in terms of obtaining false-positive results.

Conclusions Overall it is clear that the local administration of glucocorticoid has significant negative effects on tendon cells in vitro, including reduced cell viability, cell proliferation and collagen synthesis. There is increased collagen disorganisation and necrosis as shown by in vivo studies. The mechanical properties of tendon are also significantly reduced. This review supports the emerging clinical evidence that shows significant long-term harms to tendon tissue and cells associated with glucocorticoid injections.

Appendix 1 See below for Table A1. Appendix A1 Full inclusion and exclusion criteria Inclusion

Exclusion

Studies must relate to tendon tissue or tendon cells No control group following the local administration of steroid Steroid is defined as glucocorticoid Case reports, case series and review articles Animal and human tendon cells or tendon tissue Systemic and not local steroid administration In vivo and in vitro studies Studies relating to anabolic steroid Glucocorticoid treatment alone Fibroblasts not derived from tendon Not available in the English language

Appendix 2 See below for Table A2.

Appendix A2 Methodological quality assessment document (the number of “yes” answers was counted for each study to give a total score out of 8) Number

Criteria

1

Study population clearly described (animals/humans) Control group clearly described Sampling method clearly described Steroid clearly described (name/dose) Quantitative method or semi-quantitative method using minimum of 2 independent observers Validity and/or reliability of methods described Statistical significance stated for results Study limitations mentioned

2 3 4 5

6 7 8

Yes/No/Unclear

B.J.F. Dean et al. / Seminars in Arthritis and Rheumatism ] (2013) ]]]–]]]

6

Appendix 3 See below for Table A3.

Appendix A3 Summary of included studies and methodological scores References

Human/ animal

Study type

Methdological score

Akpinar et al. [59] Balasubramaniam and Prathap [58] Buck and Wilhelm [73] Dombi and Halsall [68] Gonzalez [86] Haraldsson et al. [79] Haraldsson et al. [87] Hugate et al. [80] Kapetanos [64] Kempka et al. [49]

Animal Animal

In vivo In vivo

6 3

In In In In In In In In

vivo vitro vivo vivo vitro vivo vivo vitro

3 4 4 8 7 6 6 7

Kennedy and Willis [40] Ketchum [85] Kim et al. [65] Koeke et al. [89] Lee et al. [75] Lee and Ling [56] Mackie et al. [83] Martin et al. [81] Martins et al. [78] Matthews et al. [84] McWhorter et al. [82] Mikolyzk et al. [29] Muto et al. [90] Oikarinen [69] Oikarinen [70] Oikarinen et al. [71] Oikarinen et al. [67] Oikarinen [72] Oxlund [30] Oxlund [66] Piper et al. [88] Plotkin et al. [31] Poulsen et al. [48] Poulsen et al. [77]

Animal Animal Animal Animal Animal Animal Animal Human/ animal Animal Animal Human Animal Animal Human Animal Animal Animal Animal Animal Animal Human Animal Animal Animal Animal Animal Animal Animal Animal Animal Human Human

4 7 8 8 8 5 8 8 8 6 7 8 8 5 5 5 6 4 8 7 7 7 8 8

Scutt et al. [55] Sendzik et al.[47] Shapiro et al. [62] Tatari et al. [63] Tempfer et al. [52] Tillander et al. [57] Tsai et al. [54] Tsai et al. [61] Unverferth and Olix [60] Wei et al. [74] Wong et al. [44] Wong et al. [50] Wong et al. [76] Wong et al. [45] Zargar Baboldashti et al. [46] Zhang et al. [53]

Animal Human Animal Animal Human Animal Animal Animal Animal Animal Human Human Human Human Human Human

vivo vivo vitro vivo vivo vivo vivo vivo vivo vivo vivo vivo vitro vitro vitro vitro vitro vitro vivo vivo vitro vivo vitro vivo/ in vitro In vitro In vitro In vivo In vivo In vitro In vivo In vitro In vitro In vivo In vivo In vitro In vitro In vitro In vitro In vitro In vitro

In In in In In In In In In In In In In In In In In In In In In In In In

7 8 5 4 8 5 5 5 5 6 7 7 8 8 8 8

References [1] Glyn J. The discovery and early use of cortisone. J R Soc Med 1998;91:513–7 [Epub 1999/03/10]. [2] Authority NBS. Electronic Prescribing & Financial Information for Practices (ePFIP), http://wwwnhsbsanhsuk/PrescriptionServices/963aspx; 2009. [3] Arroll B, Goodyear-Smith F. Corticosteroid injections for osteoarthritis of the knee: meta-analysis. Br Med J 2004;328:869. [4] Arroll B, Goodyear-Smith F. Corticosteroid injections for painful shoulder: a meta-analysis. Br J Gen Prac 2005;55:224–8 [Epub 2005/04/06].

[5] Staal JB, de Bie R, de Vet HC, Hildebrandt J, Nelemans P. Injection therapy for subacute and chronic low-back pain. Cochrane Database Syst Rev 2008 [CD001824, Epub 2008/07/23]. [6] Coombes BK, Bisset L, Vicenzino B. Efficacy and safety of corticosteroid injections and other injections for management of tendinopathy: a systematic review of randomised controlled trials. Lancet 2010;376:1751–67 [Epub 2010/ 10/26]. [7] Buchbinder R, Green S, Youd JM. Corticosteroid injections for shoulder pain. Cochrane Database Syst Rev 2003 [CD004016, Epub 2003/01/22]. [8] Scott A, Docking S, Vicenzino B, Alfredson H, Zwerver J, Lundgreen K, et al. Sports and exercise-related tendinopathies: a review of selected topical issues by participants of the second International Scientific Tendinopathy Symposium (ISTS) Vancouver 2012. Br J Sports Med 2013;47:536–44 [Epub 2013/04/ 16]. [9] Coombes BK, Bisset L, Brooks P, Khan A, Vicenzino B. Effect of corticosteroid injection, physiotherapy, or both on clinical outcomes in patients with unilateral lateral epicondylalgia: a randomized controlled trial. J Am Med Assoc 2013;309:461–9 [Epub 2013/02/07]. [10] Mahler F, Fritschy D. Partial and complete ruptures of the Achilles tendon and local corticosteroid injections. Br J Sports Med 1992;26:7–14 [Epub 1992/03/ 01]. [11] Shrier I, Matheson GO, Kohl HW 3rd. Achilles tendonitis: are corticosteroid injections useful or harmful? Clin J Sport Med 1996;6:245–50 [Epub 1996/10/ 01]. [12] Maffulli N, Wong J. Rupture of the Achilles and patellar tendons. Clin Sports Med 2003;22:761–76 [Epub 2003/10/17]. [13] van der Linden PD, Sturkenboom MC, Herings RM, Leufkens HM, Rowlands S, Stricker BH. Increased risk of achilles tendon rupture with quinolone antibacterial use, especially in elderly patients taking oral corticosteroids. Arch Intern Med 2003;163:1801–7 [Epub 2003/08/13]. [14] Mandel S, Schilling J, Peterson E, Rao DS, Sanders W. A retrospective analysis of vertebral body fractures following epidural steroid injections. J Bone Joint Surg 2013;95:961–4. [15] Barnes PJ. Corticosteroid effects on cell signalling. Eur Respir J 2006;27:413–26 [Epub 2006/02/03]. [16] Payne DN, Adcock IM. Molecular mechanisms of corticosteroid actions. Paediatr Respir Rev 2001;2:145–50 [Epub 2003/01/18]. [17] Lannan EA, Galliher-Beckley AJ, Scoltock AB, Cidlowski JA. Proinflammatory actions of glucocorticoids: glucocorticoids and TNFalpha coregulate gene expression in vitro and in vivo. Endocrinology 2012;153:3701–12 [Epub 2012/06/08]. [18] Dean BJ, Franklin SL, Carr AJ. The peripheral neuronal phenotype is important in the pathogenesis of painful human tendinopathy: a systematic review. Bone Joint Res 2012;1:158–66 [Epub 2013/04/24]. [19] Dean BJF, Franklin SL, Carr AJ. A systematic review of the histological and molecular changes in rotator cuff disease. BoneJoint Res 2012;1:158–66. [20] Sharma P, Maffulli N. Biology of tendon injury: healing, modeling and remodeling. J Musculoskelet Neuronal Interact 2006;6:181–90 [Epub 2006/ 07/20]. [21] Ackermann PW, Li J, Lundeberg T, Kreicbergs A. Neuronal plasticity in relation to nociception and healing of rat achilles tendon. J Orthop Res 2003;21: 432–41 [Epub 2003/04/23]. [22] Sharma P, Maffulli N. Tendon injury and tendinopathy: healing and repair. J Bone Joint Surg Am 2005;87:187–202 [Epub 2005/01/07]. [23] Rees JD, Stride M, Scott A. Tendons—time to revisit inflammation. Br J Sports Med 2013, http://www.ncbi.nlm.nih.gov/pubmed/23476034 [Epub 2013/03/12]. [24] Alfredson H, Lorentzon R. Chronic tendon pain: no signs of chemical inflammation but high concentrations of the neurotransmitter glutamate. Implications for treatment? Current Drug Targets 2002;3:43–54 [Epub 2002/03/20]. [25] Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gotzsche PC, Ioannidis JP, et al. The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration. Br Med J 2009;339:b2700 [Epub 2009/07/23]. [26] Cochrane Handbook for Systematic Reviews of Interventions. 〈http://wwwco chraneorg/training/cochrane-handbook〉; 2012. [27] Hegedus EJ, Cook C, Brennan M, Wyland D, Garrison JC, Driesner D. Vascularity and tendon pathology in the rotator cuff: a review of literature and implications for rehabilitation and surgery. Br J Sports Med 2010;44:838–47 [Epub 2009/03/19]. [28] Neyeloff JL, Fuchs SC, Moreira LB. Meta-analyses and Forest plots using a microsoft excel spreadsheet: step-by-step guide focusing on descriptive data analysis. BMC Res Notes 2012;5:52 [Epub 2012/01/24]. [29] Mikolyzk DK, Wei AS, Tonino P, Marra G, Williams DA, Himes RD, et al. Effect of corticosteroids on the biomechanical strength of rat rotator cuff tendon. J Bone Joint Surg Am 2009;91:1172–80. [30] Oxlund H. The influence of a local injection of cortisol on the mechanical properties of tendons and ligaments and the indirect effect on skin. Acta Orthop Scand 1980;51:231–8. [31] Plotkin MB, Foss ML, Goldin B, Ellis DG. Dose-response effects of antiinflammatory steroid injections on mechanical properties of rat tail tendons. Med Sci Sports 1976;8:230–4. [32] Lewis JS. Rotator cuff tendinopathy. Br J Sports Med 2009;43:236–41 [Epub 2008/09/20]. [33] Neer CS 2nd. Impingement lesions. Clin Orthop Relat Res 1983:70–7 [Epub 1983/03/01].

B.J.F. Dean et al. / Seminars in Arthritis and Rheumatism ] (2013) ]]]–]]] [34] Fredberg U, Stengaard-Pedersen K. Chronic tendinopathy tissue pathology, pain mechanisms, and etiology with a special focus on inflammation. Scand J Med Sci Sports 2008;18:3–15 [Epub 2008/02/26]. [35] Seitz AL, McClure PW, Finucane S, Boardman ND III, Michener LA. Mechanisms of rotator cuff tendinopathy: intrinsic, extrinsic, or both? Clin Biomech 2011;26:1–12. [36] Xu Y, Murrell GA. The basic science of tendinopathy. Clin Orthop Relat Res 2008;466:1528–38 [Epub 2008/05/15]. [37] Riley G. The pathogenesis of tendinopathy. A molecular perspective. Rheumatology (Oxford) 2004;43:131–42 [Epub 2003/07/18]. [38] Millar NL, Hueber AJ, Reilly JH, Xu Y, Fazzi UG, Murrell GA, et al. Inflammation is present in early human tendinopathy. Am J Sports Med 2010;38:2085–91 [Epub 2010/07/03]. [39] Abate M, Silbernagel KG, Siljeholm C, Di Iorio A, De Amicis D, Salini V, et al. Pathogenesis of tendinopathies: inflammation or degeneration? Arthritis Res Ther 2009;11:235 [Epub 2009/07/14]. [40] Kennedy JC, Willis RB. The effects of local steroid injections on tendons: a biomechanical and microscopic correlative study. Am J Sports Med 1976;4: 11–21. [41] Peters-Veluthamaningal C, van der Windt DA, Winters JC, Meyboom-de Jong B. Corticosteroid injection for trigger finger in adults. Cochrane Database Syst Rev 2009 [CD005617, Epub 2009/01/23]. [42] Peters-Veluthamaningal C, Winters JC, Groenier KH, Meyboom-DeJong B. Randomised controlled trial of local corticosteroid injections for de Quervain's tenosynovitis in general practice. BMC Musculoskelet Disord 2009;10:131 [Epub 2009/10/29]. [43] Lee DH, Han SB, Park JW, Lee SH, Kim KW, Jeong WK. Sonographically guided tendon sheath injections are more accurate than blind injections: implications for trigger finger treatment. J Ultrasound Med 2011;30:197–203 [Epub 2011/ 01/27]. [44] Wong MW, Lui WT, Fu SC, Lee KM. The effect of glucocorticoids on tendon cell viability in human tendon explants. Acta Orthop 2009;80:363–7 [Epub 2009/ 05/08]. [45] Wong MW, Tang YY, Lee SK, Fu BS, Chan BP, Chan CK. Effect of dexamethasone on cultured human tenocytes and its reversibility by platelet-derived growth factor. J Bone Joint Surg Am 2003;85–A:1914–20 [Epub 2003/10/18]. [46] Zargar Baboldashti N, Poulsen RC, Franklin SL, Thompson MS, Hulley PA. Platelet-rich plasma protects tenocytes from adverse side effects of dexamethasone and ciprofloxacin. Am J Sports Med 2011;39:1929–35. [47] Sendzik J, Shakibaei M, Schafer-Korting M, Lode H, Stahlmann R. Synergistic effects of dexamethasone and quinolones on human-derived tendon cells. Int J Antimicrob Agents 2010;35:366–74. [48] Poulsen RC, Carr AJ, Hulley PA. Protection against glucocorticoid-induced damage in human tenocytes by modulation of ERK, Akt, and forkhead signaling. Endocrinology 2011;152:503–14 [Epub 2011/01/07]. [49] Kempka G, Ahr HJ, Ruther W, Schluter G. Effects of fluoroquinolones and glucocorticoids on cultivated tendon cells in vitro. Toxicol In Vitro 1996;10: 743–54. [50] Wong MW, Tang YN, Fu SC, Lee KM, Chan KM. Triamcinolone suppresses human tenocyte cellular activity and collagen synthesis. Clin Orthop Relat Res 2004;421:277–81 [Epub 2004/05/05]. [51] Moosmayer S, Tariq R, Stiris M, Smith HJ. The natural history of asymptomatic rotator cuff tears: a three-year follow-up of fifty cases. J Bone Joint Surg Am 2013;95:1249–55 [Epub 2013/07/19]. [52] Tempfer H, Gehwolf R, Lehner C, Wagner A, Mtsariashvili M, Bauer HC, et al. Effects of crystalline glucocorticoid triamcinolone acetonide on cultered human supraspinatus tendon cells. Acta Orthop 2009;80:357–62. [53] Zhang J, Keenan C, Wang JH. The effects of dexamethasone on human patellar tendon stem cells: implications for dexamethasone treatment of tendon injury. J Orthop Res 2013;31:105–10. [54] Tsai W-C, Tang F-T, Wong M-K, Yen H-C, Pang J-HS. Decreased expression of proliferating cell nuclear antigen is associated with dexamethasone inhibition of the proliferation of rat tendon fibroblasts. J Rheumatol 2002;29(11): 2397–402. [55] Scutt N, Rolf CG, Scutt A. Glucocorticoids inhibit tenocyte proliferation and tendon progenitor cell recruitment. J Orthop Res 2006;24:173–82. [56] Lee SK, Ling CM. The response of human tendon to hydrocortisone injection. Singapore Med J 1975;16:259–62 [Epub 1975/12/01]. [57] Tillander B, Franzen LE, Karlsson MH, Norlin R. Effect of steroid injections on the rotator cuff: an experimental study in rats. J Shoulder Elbow Surg 1999;8:271–4. [58] Balasubramaniam P, Prathap K. The effect of injection of hydrocortisone into rabbit calcaneal tendons. J Bone Joint Surg Br 1972;54:729–34 [Epub 1972/11/01]. [59] Akpinar S, Hersekli MA, Demirors H, Tandogan RN, Kayaselcuk F. Effects of methylprednisolone and betamethasone injections on the rotator cuff: an experimental study in rats. Adv Ther 2002;19:194–201. [60] Unverferth LJ, Olix ML. The effect of local steroid injections on tendon. J Sports Med 1973;1:31–7. [61] Tsai WC, Tang FT, Wong MK, Pang JH. Inhibition of tendon cell migration by dexamethasone is correlated with reduced alpha-smooth muscle actin gene expression: a potential mechanism of delayed tendon healing. J Orthop Res 2003;21:265–71. [62] Shapiro PS, Rohde RS, Froimson MI, Lash RH, Postak P, Greenwald AS. The effect of local corticosteroid or ketorolac exposure on histologic and biomechanical properties of rabbit tendon and cartilage. Hand (New York, NY) 2007;2:165–72.

7

[63] Tatari H, Kosay C, Baran O, Ozcan O, Ozer E. Deleterious effects of local corticosteroid injections on the Achilles tendon of rats. Arch Orthop Trauma Surg 2001;121:333–7. [64] Kapetanos G. The effect of the local corticosteroids on the healing and biomechanical properties of the partially injured tendon. Clinical Orthop Relat Res 1982;163:170–9 [Epub 1982/03/01]. [65] Kim YS, Bigliani LU, Fujisawa M, Murakami K, Chang SS, Lee HJ, et al. Stromal cell-derived factor 1 (SDF-1, CXCL12) is increased in subacromial bursitis and downregulated by steroid and nonsteroidal anti-inflammatory agents. J Orthop Res 2006;24:1756–64. [66] Oxlund H. Long term local cortisol treatment of tendons and the indirect effect on skin. An experimental study in rats. Scand J Plast Reconstr Surg 1982;16:61–6. [67] Oikarinen A, Mäkelä J, Vuorio T, Vuorio E. Comparison on collagen gene expression in the developing chick embryo tendon and heart. Tissue and development time-dependent action of dexamethasone. Biochim Biophys Acta 1991;1089:40–6. [68] Dombi GW, Halsall HB. Collagen fibril formation in the presence of dexamethasone disodium phosphate. Connect Tissue Res 1986;15:257–68. [69] Oikarinen A. Effect of betamethasone-17-valerate on synthesis of collagen and prolyl hydroxylase activity in chick embryo tendon cells. Biochem Pharmacol 1977;26:875–9. [70] Oikarinen A. Effect of cortisol acetate on collagen biosynthesis and on the activities of prolyl hydroxylase, lysyl hydroxylase, collagen galactosyltransferase and collagen glucosyltransferase in chick-embryo tendon cells. Biochem J 1977;164:533–9. [71] Oikarinen AI, Vuorio EI, Zaragoza EJ, Palotie A, Chu ML, Uitto J. Modulation of collagen metabolism by glucocorticoids. Receptor-mediated effects of dexamethasone on collagen biosynthesis in chick embryo fibroblasts and chondrocytes. Biochem Pharmacol 1988;37:1451–62. [72] Oikarinen J. Cortisol induces (2′-5′)oligoadenylate synthetase in cultured chick embryo tendon fibroblasts. Biochem Biophys Res Commun 1982;105:876–81 [Epub 1982/04/14]. [73] Buck RC, Wilhelm DL. The influence of cortisone on the regeneration of tendon and tracheal epithelium in the rat. Br J Exp Pathol 1952;33:562–6 [Epub 1952/12/01]. [74] Wei AS, Callaci JJ, Juknelis D, Marra G, Tonino P, Freedman KB, et al. The effect of corticosteroid on collagen expression in injured rotator cuff tendon. J Bone Joint Surg Am 2006;88:1331–8. [75] Lee H-J, Kim Y-S, Ok J-H, Lee Y-K, Ha MY. Effect of a single subacromial prednisolone injection in acute rotator cuff tears in a rat model. 2013 [Epub 2013/01/31]. [76] Wong MW, Tang YY, Lee SK, Fu BS. Glucocorticoids suppress proteoglycan production by human tenocytes. Acta Orthop 2005;76:927–31 [Epub 2006/ 02/14]. [77] Poulsen RC, Watts AC, Murphy RJ, Snelling SJ, Carr AJ, Hulley PA. Glucocorticoids induce senescence in primary human tenocytes by inhibition of sirtuin 1 and activation of the p53/p21 pathway: in vivo and in vitro evidence. Ann Rheum Dis 2013 [Epub 2013/06/04]. [78] Martins CA, Bertuzzi RT, Tisot RA, Michelin AF, do Prado JM, Stroher A, et al. Dextrose prolotherapy and corticosteroid injection into rat Achilles tendon. Knee Surg Sports Traumatol Arthrosc 2012;20:1895–900. [79] Haraldsson BT, Aagaard P, Crafoord-Larsen D, Kjaer M, Magnusson SP. Corticosteroid administration alters the mechanical properties of isolated collagen fascicles in rat-tail tendon. Scand J MedSci Sports 2009;19:621–6. [80] Hugate R, Pennypacker J, Saunders M, Juliano P. The effects of intratendinous and retrocalcaneal intrabursal injections of corticosteroid on the biomechanical properties of rabbit Achilles tendons. J Bone Joint Surg Am 2004;86-A:794–801. [81] Martin DF, Carlson CS, Berry J, Reboussin BA, Gordon ES, Smith BP. Effect of injected versus iontophoretic corticosteroid on the rabbit tendon. South Med J 1999;92:600–8. [82] McWhorter JW, Francis RS, Heckmann RA. Influence of local steroid injections on traumatized tendon properties. A biomechanical and histological study. Am J Sports Med 1991;19:435–9. [83] Mackie JW, Goldin B, Foss ML, Cockrell JL. Mechanical properties of rabbit tendons after repeated anti-inflammatory steroid injections 1974;6:198–202. [84] Matthews LS, Sonstegard DA, Phelps DB. A biomechanical study of rabbit patellar tendon: effects of steroid injection. J Sports Med 1974;2:349–57. [85] Ketchum LD. Effects of triamcinolone on tendon healing and function. A laboratory study. Plast Reconstr Surg 1971;47:471–82. [86] Gonzalez RI. Experimental tendon repair within the flexor tunnels; the use of hydrocortisone without improvement of function in the dog. J Bone Joint Surg Am 1953;35-A:991–3. [87] Haraldsson BT, Langberg H, Aagaard P, Zuurmond AM, van El B, Degroot J, et al. Corticosteroids reduce the tensile strength of isolated collagen fascicles. The American journal of sports medicine 2006;34(12):1992–7. [88] Piper SL, Laron D, Manzano G, Pattnaik T, Liu X, Kim HT, et al. A comparison of lidocaine, ropivacaine and dexamethasone toxicity on bovine tenocytes in culture. The Journal of bone and joint surgery British volume 2012;94(6): 856–62 Epub 2012/05/26. [89] Koeke PU, Parizotto NA, Carrinho PM, Salate AC. Comparative study of the efficacy of the topical application of hydrocortisone, therapeutic ultrasound and phonophoresis on the tissue repair process in rat tendons. Ultrasound in medicine & biology 2005;31(3):345–50. [90] Muto T, Kokubu T, Mifune Y, Sakata R, Nagura I, Nishimoto H, Harada Y, Nishida K, Kuroda R, Kurosaka M. Platelet-rich plasma protects rotator cuffderived cells from the deleterious effects of triamcinolone acetonide. Orthop Res 2013 Jun;31(6):976–82.