CME

Intra-articular steroid injections for painful knees

Systematic review with meta-analysis

Marshall Godwin, MD, MSC, CCPF, FCFP Martin Dawes, MD, FRCGP

This article has been peer reviewed. Cet article a fait l’objet d’une évaluation externe. Can Fam Physician 2004;50:241-248.

Dr Godwin is a Professor and Director of Research in the Department of Family Medicine at Queen’s University in Kingston, Ont. Dr Dawes is Chair of the Department of Family Medicine at McGill University in Montreal, Que.

A 60-year-old man with osteoarthritis (OA) in his knee has had a lot of pain recently. You were planning to inject his knee with a corticosteroid “depo” preparation and mentioned this to a colleague. She immediately said she thought there was no evidence that intra-articular injections of corticosteroids “worked” for OA.

You have been injecting knees and other joints with steroids for many years, and your clinical impression is that it often seemed to help. Your “evidence,” however, is experiential, and you have never checked the literature for studies on this treatment. Neither has your colleague. For a 60-year-old man with painful OA of the knee, will intra-articular injection of a depo-corticosteroid preparation decrease pain without causing serious side effects?

METHODS

Literature search

We searched MEDLINE from 1966 to December 2002 using PubMed and Ovid; the Cochrane Library, including the database of systematic reviews and the register of controlled trials; and EMBASE. We also used the Internet search engine Google.com. Search terms used were osteoarthritis, knee, corticosteroid or glucocorticoids, and intra-articular or intraarticular. The search was limited to studies done on human adults, randomized controlled trials, systematic reviews, and guidelines. There were no language limits. Randomized controlled trials (RCTs) and systematic reviews were primarily sought, but articles of related interest were considered. In this first search, we found three RCTs,1-3 a nonsystematic review,4 and two reports of complications from knee injections.5,6

One RCT1 was used to seek further articles with the “Related articles” feature of PubMed. This feature does not use the limits that were set, so 161 articles were found. Of these, three were RCTs.7-9 Manual checking of references in the review article and the RCTs identified two RCTs published in 198110 and 1996,11 and several papers published in the 1950s reporting on use of non-depo formulations (hydrocortisone acetate) of steroids.12,13

We did not find a completed review on OA treatment using corticosteroid injection by the Cochrane Collaboration. There is, however, a protocol description14 in the Cochrane database.

The Internet search using Google.com found two potential sources of additional information: the American College of Rheumatology’s OA treatment guidelines15 and the EULAR’s (European League Against Rheumatism) recommendations for management of knee OA.16 No additional primary studies were found in their reference lists.

Inclusion criteria

Both RCTs and systematic reviews that looked specifically at intra-articular steroid injection for OA of the knee were considered. We included all studies that compared the newer, long-acting, potent forms of depo-corticosteroid (eg, triamcinolone hexacetonide, methylprednisolone, betamethasone, and cortivazol) with placebo. Table 11,3,7,8,11 gives information on the five studies selected. Studies done in the 1950s were not included because they were either not controlled trials or they used short-acting hydrocortisone acetate rather than the longer-acting depo formulations now available. Studies comparing two different types of intra-articular steroids were not included unless there was also a placebo arm.

Critical appraisal of selected articles

Those assessing the articles were not blinded to author or citation source, but were not acquainted with any of the authors of the articles included in the review. The critical appraisal process considered the validity of the methods, the results, and how well the results could be applied to clinical practice.17,18 Details of methods, descriptions of patients studied, and intervention and control procedures were assessed. Absolute risk reduction (ARR) and number needed to treat (NNT) for event-based outcomes were calculated for each study when sufficient data were provided.

When mean scores were used as outcomes, differences between the scores and 95% confidence intervals (CI) of these differences were calculated if standard deviations (SD) were given by the author. If SDs were not provided, P values given by the authors were used. A meta-analysis was conducted using the RevMan software available from the Cochrane Collaboration website ( http://www.cochrane.org/cochrane/revman41.htm).

For the meta-analysis, we considered both dichotomous and continuous outcomes. Target levels of pain reduction differed from study to study (Table 21,3,7,8,11); however, in combining the results, we considered only whether the target level set by the investigators had been achieved. Continuous outcomes were patients’ assessments of level of pain on a visual analogue scale (VAS) using ranges of either 0 to 10 cm or 0 to 100 mm. For the meta-analysis of VAS scores, we were not able to include the Friedman and Moore3 and Dieppe et al8 studies because SDs were not given, and, for the Jones and Doherty11 study, we estimated SDs from the interquartile ranges in the figures. Timing of measurement of outcomes was fairly consistent among studies. Four of the five studies measured outcomes at 1 week, two at 3 to 4 weeks, and two at 6 to 8 weeks. One study also measured outcomes at 12 weeks.

RESULTS

A summary of results of the five studies is shown in Table 2.1,3,7,8,11 Meta-analysis results are shown in Figure 1.1,3,7,8,11 Figure 1A shows results of a meta-analysis of the effect of intra-articular corticosteroid injection. Figure 1B shows similar results, except each bar represents scores on a VAS pain scale. Tests of heterogeneity are non-significant, indicating data can be pooled.

At 1 week, treated patients were more likely to have achieved the level of pain reduction investigators thought was clinically significant; they scored significantly lower on the VAS. At 3 to 4 weeks, achievement of target pain reduction remained significant, but the difference in VAS scores was no longer significant. At 6 to 8 weeks, there was no difference in achievement of target pain reduction or in VAS score between treatment and control groups.

None of the articles included in the review used a local anesthetic in combination with a steroid in the treatment arm, as is common in practice. This was probably because the investigators wanted to ensure they were measuring only the effects of the steroids.

As for harm, none of the investigators in the five trials reported adverse consequences of intra-articular injections, and few adverse effects are reported in the literature. Creamer4 in his 1999 review reported that iatrogenic infection occurred at a rate of 1:14 000 to 1:50 000. Evidence of accelerated deterioration of the joint due to repeated injections is very weak4-6; most authors believe it is an effect of the disease and not the injections.4,5,19 Patients report the procedure itself to be painful or very painful about 20% of the time.1

To ensure we were up-to-date, we did another MEDLINE search just before we submitted this article. We found an RCT looking at the safety and efficacy of triamcinolone over the long term published in February 2003.20 Patients were given either 40 mg of triamcinolone or saline into affected knees every 3 months for 2 years. They received a total of eight injections each. Pain scores improved in both groups during the 2-year period and were similar in both groups at 1 year and 2 years. Range of motion of the knee was better in the triamcinolone group at 1 year but not at 2 years. Treated patients had less pain at night and marginally less stiffness in knees at 2 years.

The importance of this study is that radiologic examination did not show any difference in deterioration in the triamcinolone group compared with placebo. Also, there were no adverse local or systemic affects. This study has not been included in the meta-analysis because it used multiple injections and did not measure outcomes in the short term (1 to 12 weeks). Its purpose was to evaluate the long-term effect of multiple injections in the knee. We include it here for the sake of completeness and because it reinforces our conclusion that adverse effects are rare.

DISCUSSION

Treatment effects were consistent among the five studies (Table 21,3,7,8,11). The four studies that measured effect at 1 week show a significant reduction in pain (assessed on a VAS) compared with placebo. Three of the studies showed an effect at 1 week when patients’ subjective assessments of pain relief were used or when a predetermined clinically significant level of pain reduction was used. No study showed an effect of triamcinolone beyond 1 week. Methylprednisolone, however, showed a continuing effect at 3 weeks, and cortivazol at 4 weeks. The meta-analysis confirms the general impression of the study results: that intra-articular steroid injections are useful in reducing pain for up to 4 weeks.

Clinical application

These studies were all performed in a secondary care environment by rheumatologists, who also gave the injections. Patients all had moderate-to-severe OA. Before treating patients with intra-articular steroids, primary care physicians should consider whether they are comfortable and experienced in doing intra-articular injections and whether patients’ disease is similar to that of people who have been shown to benefit from this treatment. Pain relief is likely to be short lived; it will likely last for a week and probably for a month, but not beyond that. A short-term effect can sometimes be useful (eg, to control severe pain while waiting for nonsteroidal anti-inflammatory drugs [NSAIDs] to work, in situations where patients need rapid and substantial relief of pain for an upcoming activity, and when pain is affecting health due to sleep interruption). Osteoarthritic patients who have gastrointestinal side effects from NSAIDs might well be good candidates for steroid injection.

Alternative interpretations of the data

It is likely that the effect of treatment is real because of the consistency of results from study to study. The studies overall appear well done, especially the recent ones. The effect seems short lived, however, and in several situations, CIs suggest that intra-articular steroids could increase pain for some people. On balance, there seems a greater likelihood of benefit than of harm within the first month; but then a decreasing effect over time could lead to increased pain. Patients’ informed choice will be important in making the decision to use intra-articular steroids.

Limitations

We believe the results of this meta-analysis provide a valid assessment of the effect of intra-articular steroids for treatment of knee OA. The following methodologic limitations should be kept in mind, however. We did not contact experts in the field to seek out unpublished data; assessments were done by only two reviewers who were not blinded to the source of the articles; and the five studies selected used slightly different end points or targets for successful pain reduction (we accepted these targets and combined them in a single outcome of successful or failed treatment). Differences in effect were not large.

Bottom line

This systematic review with meta-analysis supports the recommendations of American19 and international20 authorities that intra-articular injection of corticosteroids provides short-term relief of the pain of OA of the knee. Intra-articular corticosteroids result in a clinically and statistically significant reduction in knee pain 1 week after injection that continues for 3 to 4 weeks. Adverse events were rarely reported. The procedure should be used for patients with moderate-to-severe OA. Physicians using this treatment should be aware of its limitations. Patients should be informed of the benefits and possible adverse effects of intra-articular injection, and decisions on whether to inject should be made jointly by physicians and patients.

Competing interests

None declared

Correspondence to: Dr Marshall Godwin, Centre for Studies in Primary Care, Department of Family Medicine, 220 Bagot St, PO Bag 8888, Kingston, ON K7L 5E9; telephone (613) 549-4480; e-mail godwinm@post.queensu.ca

References

1. Ravaud P, Moulinier L, Giraudeau B, Ayral X, Guerin C, Noel E, et al. Effects of joint lavage and steroid injection in patients with osteoarthritis of the knee: results of a multicenter, randomized, controlled trial. Arthritis Rheum 1999;42:475-82.
2. Sambrook PN, Champion GD, Browne CD, Cairns D, Cohen ML, Day RO, et al. Corticosteroid injection for osteoarthritis of the knee: peripatellar compared to intra-articular route. Clin Exp Rheumatol 1989;7(6):609-13.
3. Friedman DM, Moore ME. The efficacy of intraarticular steroids in osteoarthritis: a double-blind study. J Rheumatol 1980;7(6):850-6.
4. Creamer P. Intra-articular corticosteroid treatment in osteoarthritis. Curr Opin Rheumatol 1999;11(5):417-21.
5. Hunter JA, Blyth TH. A risk-benefit assessment of intra-articular corticosteroids in rheumatic disorders. Drug Saf 1999;21(5):353-65.
6. Wada J, Koshino T, Morii T, Sugimoto K. Natural course of osteoarthritis of the knee treated with or without intraarticular corticosteroid injections. Bull Hosp Joint Dis 1993;53(2):45-8.
7. Gaffney K, Ledingham J, Perry JD. Intra-articular triamcinolone hexacetonide in knee osteoarthritis: factors influencing the clinical response. Ann Rheum Dis 1995;54:379-81.
8. Dieppe P, Cushnaghan J, Jasani MK, McCrae F, Watt I. A two-year, placebo-controlled trial of non-steroidal anti-inflammatory therapy in osteoarthritis of the knee joint. Br J Rheumatol 1993;32(7):595-600.
9. Creamer P, Hunt M, Dieppe P. Pain mechanisms in osteoarthritis of the knee: effect of intraarticular anesthetic. J Rheumatol 1996;23(6):1031-6.
10. Valtonen EJ. Clinical comparison of triamcinolone hexacetonide and betamethasone in the treatment of osteoarthrosis of the knee-joint. Scand J Rheumatol Suppl 1981;41:1-7.
11. Jones A, Doherty M. Intra-articular corticosteroids are effective in osteoarthritis but there are no clinical predictors of response. Ann Rheum Dis 1996;55:829-32.
12. Hollander JL. Intra-articular hydrocortisone in arthritis and allied conditions; a summary of two years’ clinical experience. J Bone Joint Surg Am 1953;35-A(4):983-90.
13. Miller JH, White J, Norton TH. The value of intra-articular injections in osteoarthritis of the knee. J Bone Joint Surg Br 1958;40B:636-43.
14. Bellamy N, Campbell J, Wells G, Bourne R. Intra-articular corticosteroids for osteoarthritis of the knee (Protocol for a Cochrane review). In: The Cochrane Library. Oxford, Engl: Update Software; 2000. Issue 4.
15. Subcommittee on Osteoarthritis Guidelines. Recommendations for the medical management of osteoarthrits of the hip and knee. Arthritis Rheum 2000;43:1905-15.
16. Pendleton A, Arden N, Dougados M, Doherty M, Bannwarth B, Bijlsma JW, et al. EULAR recommendations for the management of knee osteoarthritis: report of a task force of the Standing Committee for International Clinical Studies Including Therapeutic Trials (ESCISIT). Ann Rheum Dis 2000;59:936–44.
17. Guyatt GH, Sackett DL, Cook DJ. Users’ guides to the medical literature. II. How to use an article about therapy or prevention. A. Are the results of the study valid? Evidence-Based Medicine Working Group. JAMA 1993;270:2598-601.
18. Guyatt GH, Sackett DL, Cook DJ. Users’ guides to the medical literature. II. How to use an article about therapy or prevention. B. What were the results and will they help me in caring for my patients? Evidence-Based Medicine Working Group. JAMA 1994;271:59-63.
19. Balch HW, Gibson JM, El-Ghobarey AF, Bain LS, Lynvh MP. Repeated corticosteroid injections into knee joints. Rheumatol Rehabil 1977;16(3):137-40.
20. Raynauld JP, Buckland-Wright C, Wald R, Choquette D, Haraoui B, Martel-Pelletier J, et al. Safety and efficacy of long-term intraarticular steroid injections in osteoarthritis of the knee. A randomized, double-blind, placebo-controlled trial. Arthritis Rheum 2003;48:370-7.