Corticosteroids In Osteoarthritis Therapy: Scarce Evidence For A Superior Drug And Dose
Overview
The objective of this study was to measure and compare the clinical significance of various intra-articular corticosteroids (CS) in treating osteoarthritis (OA).
In October 2023, a comprehensive search was conducted using PubMed, Cochrane, and Web of Science databases, adhering to the PRISMA guidelines. The inclusion criteria encompassed animal or human randomized controlled trials (RCTs) in English, with no time restrictions, that compared different intra-articular CS for OA treatment. The quality of the articles was evaluated using the Cochrane RoB2 and GRADE guidelines for human RCTs and SYRCLE’s tool for animal RCTs.
The selection process yielded eighteen RCTs (16 human and 2 animal studies), involving 1577 patients (1837 joints) and 31 animals (51 joints). The corticosteroids analyzed included triamcinolone (14 human and 2 animal studies), methylprednisolone (7 human and 1 animal study), betamethasone (3 human studies), and dexamethasone (1 human study). The majority of the studies focused on knee OA, with three human studies and one animal study examining other joints. A meta-analysis comparing methylprednisolone and triamcinolone in humans with knee OA was conducted, evaluating VAS pain scores at various intervals: very short-term (≤2 weeks), short-term (>2 and ≤4 weeks), mid-term (>4 and ≤8 weeks), long-term (>8 and ≤12 weeks), and very long-term (>12 and ≤24 weeks). Results indicated that triamcinolone had superior post-injection outcomes compared to methylprednisolone at the very short-term interval (p = 0.028). However, no significant differences in VAS improvement were observed at other follow-up points.
The preclinical and clinical literature currently offers limited evidence for comparing different corticosteroids, making it challenging to determine the optimal drug from many available corticosteroids and to find the ideal dosage for intra-articular injections in OA treatment.
Introduction
Osteoarthritis (OA) ranks among the most prevalent joint diseases and is a significant source of pain and disability in older adults, placing a substantial burden on healthcare systems globally. Initially considered a primarily degenerative condition, OA is now understood to have a critical inflammatory component in its pathogenesis, characterized by alterations in the synovial membrane and the release of pro-inflammatory cytokines within the joint environment. These inflammatory elements trigger chondrocytes to produce degradative enzymes that affect the articular surface.
In light of this, intra-articular corticosteroid (CS) injections, including betamethasone, dexamethasone, methylprednisolone, and triamcinolone, have been proposed as a therapeutic strategy for OA management due to their anti-inflammatory and analgesic properties. Corticosteroids operate by binding to nuclear steroid receptors and disrupting the inflammatory and immune processes associated with OA, thus reducing pro-inflammatory and pain mediators. This mechanism has shown beneficial outcomes in both animal models and clinical settings, leading to the widespread adoption of corticosteroids injections in clinical practice. These injections have demonstrated effectiveness in alleviating pain and enhancing joint function in OA patients.
Numerous scientific societies and healthcare organizations, such as the Osteoarthritis Research Society International (OARSI), the American College of Rheumatology (ACR), the European Society for Clinical and Economic Aspects of Osteoporosis, Osteoarthritis and Musculoskeletal Diseases (ESCEO), and the American Academy of Orthopaedic Surgeons (AAOS), recommend intra-articular corticosteroids injections as part of a comprehensive OA treatment plan. However, these guidelines do not specify the optimal type and dosage of corticosteroids.
This systematic review and meta-analysis aim to evaluate all existing animal studies and clinical trials comparing various intra-articular corticosteroids types and dosages to determine the most effective approach for the intra-articular treatment of OA.
Method
A systematic review was conducted to compare various intra-articular corticosteroids (CS) types and dosages for osteoarthritis (OA) treatment. This study, registered on the International Prospective Register of Systematic Reviews (PROSPERO, registration number CRD42023466164), involved a literature search across PubMed, Cochrane, and Web of Science databases on October 30, 2023. The search strategy included terms related to steroids and their intra-articular application for OA treatment, with no restrictions on publication date.
Following the PRISMA guidelines, two authors (A.Be. and M.S.) independently handled the article selection and data extraction. Initial screenings of titles and abstracts were conducted using specific inclusion criteria: randomized controlled trials (RCTs) in animals or humans, published in English, comparing different intra-articular corticosteroids types and dosages for OA treatment. Exclusions encompassed non-randomized studies, non-English articles, reviews, expert opinions, preclinical in vitro studies, and studies unrelated to OA or treatment outcomes. Full-text screenings further refined the selected articles. Discrepancies were resolved by a third author (A.Bo.), and relevant data were extracted independently by A.Be. and M.S.
Data from animal studies included publication details, animal characteristics, OA model, follow-up time, corticosteroids type and dosage, injection number, morphological and histological findings, biomarkers, clinical outcomes, and adverse events. Clinical studies provided information on publication details, patient demographics, joint evaluated, follow-up time, corticosteroids type and dosage, injection number, OA grade, clinical outcomes, and adverse events. These data were compiled for analysis.
Clinical benefit scores of different corticosteroids types and dosages were considered for meta-analysis if at least three studies contributed to each time point. Outcome analyses were categorized into five follow-up periods: very short-term (≤2 weeks), short-term (>2 to ≤4 weeks), mid-term (>4 to ≤8 weeks), long-term (>8 to ≤12 weeks), and very long-term (>12 to ≤24 weeks). Results were evaluated based on their minimal clinically important difference (MCID) values.
Risk of bias and quality of evidence were assessed independently by A.Be. and M.S., with consensus reached through discussion with A.Bo. The SYRCLE tool was used for animal studies, assessing selection, performance, detection, attrition, reporting, and other biases, with ratings of ‘yes’ (low risk), ‘no’ (high risk), or ‘unclear’ (unclear risk). For clinical studies, the Cochrane RoB 2 tool was employed, examining trial design, conduct, and reporting through a series of signaling questions, resulting in judgments of ‘low’ or ‘high’ risk of bias, or expressing ‘some concerns’. The GRADE guidelines were used to evaluate the overall quality of evidence for each meta-analysis outcome, considering factors such as risk of bias, result inconsistency, evidence indirectness, imprecision, and publication bias, with the quality rated as ‘high’, ‘moderate’, ‘low’, or ‘very low’.
Statistical Analysis
The statistical analysis and Forest plotting were conducted following the methodology of Neyeloff et al. using the Meta XL tool for Microsoft Excel, performed by an independent professional statistician. The analysis employed random effects (DerSimonian & Laird) to calculate the weighted mean differences (MD) for continuous variables. Initially, heterogeneity was assessed using the Cochran Q statistic and the I2 metric, with significant heterogeneity defined as I2 ≥ 25%. For I2 < 25%, indicating no heterogeneity, a fixed effect model was used to estimate the expected values and 95% confidence intervals (CIs). If heterogeneity was present, a random-effect model was applied, and the I2 metric was reevaluated to ensure the correction of heterogeneity. A significance level of p = 0.05 was used. Group comparisons were made based on the analysis of variance between basal and follow-up scores MD.
Result
After removing duplicates, the initial search yielded 5704 records. Abstracts were screened using inclusion and exclusion criteria, resulting in 39 articles being reviewed for eligibility. Following full-text evaluation, 21 studies were excluded: 12 did not compare different corticosteroids (CS), 5 were not randomized controlled trials (RCTs), 2 were unrelated to osteoarthritis (OA), and 2 did not report treatment outcomes. Thus, 18 RCTs were included in this systematic review and meta-analysis. This included 2 animal studies (Table 1) and 16 human studies (Table 2). The first study comparing different CS was published in 1981, with an increasing publication trend peaking between 2016 and 2020. Both animal studies lacked blinding details for assessors. Thirty-one animals and 51 joints were assessed. One study evaluated 11 male guinea pigs with unilateral knee OA induced by sodium iodoacetate, treated with triamcinolone hexacetonide. The 0.4 mg/kg dose was more effective than the 0.04 mg/kg dose in preserving the articular surface, proteoglycans, reducing chondrocyte loss, and osteophyte formation. The other study involved 20 dogs with naturally occurring bilateral hip OA, treated with triamcinolone acetonide or methylprednisolone acetate. No significant differences were found between the treatments regarding pain outcomes.
Among human RCTs, nine were double-blind, one was single-blind (assessor), one was not blinded, and five did not report blinding details. These studies included 1576 patients (64.1% women, 35.9% men, mean age 63.3 years, mean BMI 29.4 kg/m²) and analyzed 1837 joints. Fourteen RCTs focused on knee OA, two on shoulder OA, and one on hip OA. Various corticosteroids types were used, including triamcinolone (14 studies, 1286 patients), methylprednisolone acetate (7 studies, 403 patients), betamethasone (3 studies, 90 patients), and dexamethasone palmitate (1 study, 24 patients). Triamcinolone was administered in acetonide form in nine studies and hexacetonide form in five studies. It was given as a crystalline suspension in eight studies and extended-release formulation in six studies. Betamethasone was used in three different forms across studies. CS doses ranged from 3 mg of betamethasone disodium phosphate to 80 mg of triamcinolone acetonide. All studies administered a single CS injection, with injected volumes reported in 15 studies, ranging from 1 to 5 mL. Lidocaine was combined with CS in four studies.
The most used clinical outcome measures included the Visual Analogue Scale (VAS) for pain in 12 studies, Western Ontario and McMaster Universities Arthritis Index (WOMAC) in seven studies, Knee Injury and Osteoarthritis Outcome Score (KOOS) in two studies, OMERACT-OARSI criteria in two studies, Lequesne index in two studies, and average daily pain (ADP) in one study. Some studies also assessed activity index and knee pain/joint tenderness.
A total of 457 adverse events (24.9%) were reported across the clinical studies. Most (437, 95.6%) were mild, such as arthralgia, headache, and back pain. Serious adverse events (20, 4.4%) were reported in five RCTs, with nine RCTs reporting no serious events, and two RCTs not reporting adverse events. Nineteen serious adverse events occurred in patients receiving triamcinolone acetonide, with the extended-release formulation having a higher rate (2.2%) compared to the crystalline suspension (1.2%). Only one serious adverse event, considered treatment-related, was reported in the methylprednisolone group.
The meta-analysis compared VAS pain outcomes between triamcinolone and methylprednisolone in knee OA at various follow-up periods. At very short term (≤2 weeks), a significant difference favored triamcinolone (p=0.028), but the difference did not exceed the minimal clinically important difference (MCID) of 1.37. No significant differences were found in VAS improvement from baseline, or at short, mid, long, and very long-term follow-ups.
Risk of bias was assessed using the SYRCLE tool for animal studies, with most items (80%) unclear and 20% at low risk. For human studies, the RoB 2 tool identified four studies with low risk, nine with some concerns, and three with high risk. The GRADE evaluation indicated low evidence quality for VAS pain analysis at all follow-ups.
Conclusion
The primary finding of this systematic review is that the current preclinical and clinical literature offers limited comparative data on different corticosteroids (CS) for intra-articular injections in osteoarthritis (OA), which obstructs identifying the optimal corticosteroids type and dosage for this treatment. Despite the widespread use of intra-articular CS for OA, there is no consensus on the most appropriate corticosteroids type or dose for clinical application.
For decades, intra-articular CS injections have been utilized to manage symptomatic OA by leveraging their anti-inflammatory and immunomodulatory properties to mitigate joint inflammation. Various CS products and dosages are commonly used in clinical practice, making it crucial to determine the best combination to maximize treatment efficacy and minimize adverse effects. Preclinical studies on CS effects on cartilage, often conducted on healthy models, have yielded inconsistent results. Nakazawa et al. noted that repeated injections of hydrocortisone acetate, triamcinolone acetonide, and dexamethasone acetate induced cartilage degradation and chondrocyte apoptosis in both in vitro and in vivo models. Similarly, Doyle et al. and Chunekamrai et al. observed adverse effects such as proteoglycan catabolism, chondrocyte necrosis, and collagen synthesis reduction in healthy equine cartilage treated with methylprednisolone acetate. Further studies also highlighted the detrimental impact of CS, particularly triamcinolone and methylprednisolone, on human chondrocyte viability in vitro.
Controlled studies, including those on large animal models, have aimed to elucidate both therapeutic and adverse effects of various CS products. Triamcinolone acetonide has shown beneficial effects in some models, whereas methylprednisolone acetate has demonstrated significant deleterious effects, leading to its discontinued use in racehorses due to welfare concerns. However, the systematic review highlighted a significant gap in research, with only two randomized controlled trials (RCTs) addressing CS comparisons, limiting the ability to draw definitive conclusions.
Clinical evidence on different CS types and dosages for OA management predominantly focuses on the knee, with 14 RCTs, while only a few studies address other joints like the shoulder and hip. This uneven distribution of research hinders the development of standardized treatment protocols, leaving clinical decisions largely to practitioners’ discretion.
A wide range of CS options is available, including dexamethasone, betamethasone, hydrocortisone, prednisolone, triamcinolone, and methylprednisolone. Among these, triamcinolone and methylprednisolone are most frequently used and studied. Current evidence suggests equivalent pain relief between these two for knee OA, though further studies with larger patient samples are needed to confirm these findings.
Extended-release formulations of triamcinolone acetonide have shown promise in reducing systemic exposure and extending therapeutic effects, but independent studies are necessary to validate these results. The review also highlighted the need for more research on optimal CS dosages, as current data is insufficient.
Safety remains a crucial concern, with some studies indicating potential long-term cartilage damage from repeated CS injections. However, there is no conclusive evidence favoring one CS product over others in terms of safety. More comprehensive trials are needed to assess long-term effects and compare different CS products and dosages.
In summary, the limited and inconsistent literature on CS comparisons impedes definitive recommendations for the best CS type and dosage for intra-articular injections in OA treatment. Future research should focus on broader comparative studies, standardized protocols, and long-term safety evaluations to guide clinical practice more effectively.
Oncology Related Tools
Other
Latest Research
