The first synthesized corticosteroid known as compound E, now known as cortisone, was administered to a patient with rheumatoid arthritis in 1948. The patient was free of pain within a few days of treatment, so the cortisone was continued. After about a month of therapy, the patient was admitted to a psychiatric ward after experiencing a variety of central nervous system adverse effects such as depression and psychosis from the cortisone therapy. The cortisone therapy was stopped, but the patient died several years later due to complications from adrenocorticotropic hormone therapy.1 This case illustrates the marked clinical improvement that can occur within a few days of receiving a corticosteroid and the devastating adverse effects that can occur with prolonged therapy.
Corticosteroids exhibit diverse genomic and nongenomic mechanisms of action on immune cells, tissue, and organs that gives them efficacy for a wide range of indications including endocrine disorders, rheumatic disorders, collagen diseases, dermatologic diseases, allergic states, ophthalmic disease, neoplastic disease, certain edematous states, and nervous system disorders. The diverse mechanisms of action that allow for the treatment of these disorders also accounts for the diverse adverse effect profile of corticosteroids.2 Every organ system in the body may be negatively impacted by long-term corticosteroid therapy: skin (eg, thinning, bruising), muscle (eg, wasting, cramping), skeletal (eg, osteoporosis), nervous (eg, psychiatric and sleep disturbances), circulatory (eg, hypertension), lymphatic (eg, lymphedema), respiratory (eg, infection), endocrine (eg, diabetes), reproductive (eg, fertility), urinary/excretory (eg, infection, glycosuria), and digestive (eg, ulcers). These adverse effects are dose and duration dependent. Although dose and duration thresholds are not well-established, the more consequential adverse effects of corticosteroids usually occur with supraphysiologic doses administered for more than 7 to 10 days.3 A physiologic or low dose of corticosteroid is defined as a prednisone equivalent dose 7.5 mg or less daily. Supraphysiologic doses are defined as medium dose greater than 7.5 mg but 30 mg or less daily, high dose greater than 30 mg but 100 mg or less daily, very high dose greater than 100 mg daily, and pulse therapy 250 mg or greater a day for at least 1 day.4
The corticosteroid regimens for patients with inflammatory disorders such as rheumatic diseases typically use higher initial doses to arrest or reverse the disease process and achieve symptomatic control. When the latter is achieved, a tapering regimen is implemented that decreases the dose of corticosteroid as rapidly as possible while maintaining disease control to minimize corticosteroid-related adverse effects. The objective of this systematic review was to assess clinical outcomes (efficacy and adverse effects) of differing tapering regimens after medium- to high-dose (>7.5 mg but ≤100 mg oral prednisone equivalent daily), extended-duration (at least 10 days) corticosteroid therapy in adults with rheumatic disorders. The focus was on randomized controlled trials (RCTs), case-control studies, and prospective observational studies that compared at least 2 tapering regimens of medium- to high-dose, extended-duration corticosteroid monotherapy in adults.
In conjunction with clinical expertise from 2 of the authors, an experienced health sciences librarian with extensive training and involvement in systematic reviews developed a comprehensive search strategy. This electronic search involved 7 bibliographic databases from inception up to June 27, 2018, including Medline/PubMed (NLM), Embase (Elsevier), Cochrane, International Pharmaceutical Abstracts, Web of Science, Scopus, and Global Index Medicus. In addition, the American College of Rheumatology Web site (www.rheumatology.org/Practice-Quality/Clinical-Support/Clinical-Practice-Guidelines), gray literature (ClinicalTrials.gov, National Guideline Clearinghouse [guideline.gov], Dissertations and Theses), citations, and a general Web search (Google and Google Scholar) were searched for the same period. No limits on language or publication status were applied. A combination of controlled vocabulary and keywords were used to search the databases. These terms included polymyalgia rheumatica (PMR) and giant cell arteritis (GCA), oral corticosteroids, and dose regimen. To identify PMR and GCA, terms such as polymyalgia arteritica, rheumatic polymyalgia, and cranial arteritis were used, whereas terms such as tapering, dosage decrease, and drug dose schedule were used to identify dosage regimens. A variety of oral corticosteroid drug names were used to make the search as all-encompassing as possible. The search strategy used is outlined in Appendix 1, http://links.lww.com/RHU/A135. This review was performed and reported to maintain consistency with the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-analyses).5 The protocol for this systematic review has not been previously published.
All studies identified in the literature search underwent title screening to determine which may potentially address the research question. Two authors independently reviewed the abstracts for relevant studies, and full texts were retrieved for those studies that appeared to fit the eligibility criteria for inclusion in this review, as well as those studies that did not have the information of interest in the abstract necessary to determine eligibility for inclusion. Disagreements regarding inclusion eligibility were resolved through discussion. To be included, a study had to report at least one efficacy parameter such as maintenance of disease remission and at least one adverse effect parameter, and the indication had to be for a rheumatic disorder. Studies were excluded if they did not include or describe the tapering regimen, if the corticosteroid was being used adjunctively with another pharmacologic therapy, if the tapering regimen was used to assess the effect of a concomitant medication, if the corticosteroid was stopped abruptly without a taper, if the corticosteroid was being used in patients younger than 18 years, if the steroid was not being administered systemically (eg, topically), if the dose at initiation of taper was 7.5 mg or less or greater than 100 mg oral prednisone equivalent, if the steroid had been used for less than 10 days, or if the corticosteroid was being used for immunosuppressive therapy.
The following methodological information was obtained by 2 authors independently using a standardized data extraction form developed specifically for this review: study design, patient demographics, and inflammatory condition being treated; interventions including corticosteroid product, initial dosing, and tapering regimen; and efficacy and adverse effect outcomes reported.
Risk of Bias Assessment
Two authors independently assessed the risk of bias of each study using the Cochrane Collaboration tool for RCTs.6 This tool assessed 6 potential domains for bias, including selection, performance, detection, attrition, reporting, and other bias. Bias was evaluated as high, low (risk of bias is low and unlikely to change results), or unclear. Disagreements were resolved through discussion.
Figure 1 details the study selection for this review. Of the 2245 citations identified through the literature search after the removal of duplicates, 234 abstracts were screened, 19 full text papers were reviewed, and 2 met criteria for inclusion for qualitative synthesis.7,8 Of the 17 full text papers that were excluded, the reasons were as follows: no comparison of tapers (n = 15), steroids not used as monotherapy for disease control (n = 1), and dose before taper initiation not within the medium- to high-dose range (n = 1).
Of the 2 studies included in the systematic review, one study involved corticosteroid removal for GCA,7 whereas the other addressed PMR.8 Although neither study compared tapering regimens of the same corticosteroid, one compared tapering strategies of methylprednisolone with prednisone8 and the other compared tapering modified release prednisone with immediate release prednisolone.7 One study was conducted in the United Kingdom,7 and the other was conducted in Italy.8
The total number of patients in each study was 12 and 50, and the mean age of patients included in both studies was over 70 (1 study reported a mean patient age of 80 years old).7,8 Both studies reported a number of different efficacy and safety outcomes, which are outlined in Table 1, but these reported outcomes were not the same from study to study. One study was a feasibility study with a duration of 26 weeks,7 whereas the other had its final follow-up point at 12 months. Throughout both studies, C-reactive protein was followed as an indicator of inflammatory improvement.
In the study by Raine et al, included patients had to be newly diagnosed with GCA within the past 4 weeks, not previously have been exposed to disease-modifying antirheumatic drugs (DMARDs) or biologic therapy, and have not responded to high doses of steroids in the past 5 days. These patients were referred from a variety of settings, including clinics, the emergency department, or the hospital.7 In the study by Viapiana et al, included patients had to have been referred to the rheumatology clinic for newly diagnosed PMR.8Table 1 further outlines the characteristics of the included studies.
Results of Individual Studies
In the study by Raine and colleagues, patients with GCA were given tapers (described in detail in Table 1) of modified release prednisone or immediate release prednisolone. After 26 weeks, 85.7% (n = 6) of the patients in the modified release prednisone group had experienced persistent disease control, whereas 80% (n = 4) of those in the prednisolone group had achieved this outcome.7 Disease control was defined as not exhibiting signs and symptoms of active disease (flare), which would be defined by an elevation in erythrocyte sedimentation rate (ESR) by more than 30 mm/h or CRP greater than 10 mg/L plus at least 1 clinical feature.7
In the study by Viapiana et al, 100% of the patients in the methylprednisolone group and 89% (23/26) of the patients in the prednisone group achieved clinical and biochemical remission.8 In addition, time to successfully taper from one dose to another was measured. The first dose change took a mean of 15.2 days in the methylprednisolone group, and 20.3 days in the prednisone group, which was the only dose change that was statistically significant. The authors noted that the longer mean was driven by the 3 patients in the prednisone group who achieved remission after a total of 26 to 49 days.8 Data extracted are reported in Table 1, and the outcomes of interest are reported in Table 2. The additional outcome data that has not been reported in Table 2 can be found in Appendix 2, http://links.lww.com/RHU/A136.
Adverse Effects Data
In the study by Raine et al, sleep was measured using the Leeds Sleep Evaluation score. From weeks 10 to 26, patients in the modified release prednisone group had poorer sleep scores than those in the prednisolone group (37.2 vs 61.9, respectively).7 Neither study was powered to detect a difference in adverse effect outcomes due to small sample sizes, but the descriptions of the outcomes are outlined in detail in Table 2.7,8 Of note, although no osteoporotic fractures were observed in the study by Viapiana et al, all patients were placed on bisphosphonate prophylaxis.8
Risk of Bias in Included Studies
The risk of bias assessment is reviewed in Table 3. Because both of the randomized trials were open label and not blinded, they were rated as having high performance bias. The study by Viapiana et al was determined to have high detection bias due to its open label nature to the investigators; however, the Raine study blinded evaluators to reduce potential bias when assessing outcomes, so this article was deemed to have low detection bias. In addition, not all prespecified secondary outcomes in the Raine study were reported (eg, time to first flare), so this study was rated as having high reporting bias. Selection bias for Viapiana was rated as unclear because the process by which patients were randomized, aside from being randomized at a 1:1 ratio, was not fully described. All other bias domains were rated as low.
Although there are 2 studies that met inclusion for this review, there are no adequately powered, long-term studies that have evaluated the optimal corticosteroid taping regimen for various rheumatic disorders despite almost 70 years of clinical use, so recommendations are based on expert opinion and retrospective and prospective case series.
The 2 major rheumatic conditions that continue to use medium- to large-dose, extended-duration corticosteroid regimens as first-line monotherapy are PMR and GCA. The tapering regimens recommended for these conditions in systematic reviews and clinical practice guidelines are based on expert opinion and retrospective and prospective case series published after the widespread use of corticosteroid treatment regimens beginning in the early 1950s.9–12 A summary of the corticosteroid tapering-related recommendations from recent systematic reviews and clinical practice guidelines are listed in Table 4.13–18 One prospective study examining daily versus alternate daily equivalent corticosteroid dosing in GCA was identified, which suggested that extended dosing intervals to alternate day regimens does not control disease as well as single or split daily doses.18 However, this study was not eligible for inclusion based on the PICOS question because it, like many others, did not provide information on tapering strategies.
Further, the introduction of a variety of novel therapies for inflammatory disorders such as rheumatologic disease has relegated the use of corticosteroids to adjunctive therapy with a first-line DMARD or an immunosuppressive agent, or as second- or third-line agents for patients with uncontrolled or relapsed disease. For example, guidelines by the American College of Rheumatology recommend primary immunosuppression with cyclophosphamide or mycophenolate mofetil for patients with more severe forms of lupus nephrtitis.19 High-dose, extended-duration corticosteroids are recommended as adjunctive therapy, but the guidelines specifically state there is insufficient data to recommend a taper regimen. In patients with rheumatoid arthritis, the current clinical practice guideline by the American College of Rheumatology provides a conditional recommendation (low quality evidence) for the possible use of low-dose, short-duration corticosteroids only when disease persists despite moderate- to high-dose DMARDs or biologic therapies.20 In patients with active ankylosing spondylitis, guidelines by the American College of Rheumatology have a strong recommendation against the use of corticosteroids.21 Monotherapy with moderate- to high-dose corticosteroids remains the current initial treatment approach to patients with PMR and GCA, but treatment alternatives such as tocilizumab for the latter disorders are currently showing promise, so the primacy of corticosteroid therapy for these exceptions may be changing in the near future.22
The introduction of novel therapies such as biologics has shifted the treatment approach for many rheumatic conditions. Corticosteroids, when used, are often administered as adjunctive therapy in lower doses and shorter courses. For example, a systematic review of randomized trials using tapering regimens of corticosteroids for rheumatoid arthritis found only 3 studies that used initial doses of prednisone equivalents greater than 7.5 mg.23 Corticosteroid tapers were used, but the tapering regimen was not the focus of the studies. Two of the studies investigated the addition of prednisolone to DMARDs, and a third investigated the effects of corticosteroid therapy on inflammation, bone density, and feasibility of steroid withdrawal.24–26
There are some obvious limitations to this review. The restriction to longitudinal studies may have eliminated some beneficial case reports and case series. The Raine study, a feasibility trial evaluating tapering regimens in GCA, took place over a 26-week period, which is much shorter than typical GCA treatment.7 In addition, the studies that exist compare tapering regimens of different pharmacologic agents—not different doses of the same agent.7,8 Although one study suggests that patients taking prednisolone may have fewer sleep disturbances than those taking a modified release formulation of prednisone over time, the small number of patients (n = 12) in this study make it difficult to strictly apply to practice.7
The search for alternatives to corticosteroids for rheumatic disorders is not surprising given the potential adverse effects of these agents on every organ system in the body, particularly when used in moderate to high doses for extended periods. A full review of corticosteroid complications and ways to prevent or ameliorate their occurrence is beyond the scope of this review, but extensive recommendations are available.3
Based on the findings of this systematic review, it is clear that a number of uncertainties remain regarding ideal corticosteroid tapering strategies. The studies in this systematic review each include tapering strategies of 2 different corticosteroid formulations. However, the question remains regarding how often doses should be changed and at what dose increments. The ideal study to answer this question would involve randomization of tapering strategies for patients with GCA or PMR using the same formulation of corticosteroid (eg, prednisone) with either (a) differing durations at each dose or (b) differing increments of dose reductions. Studies with such methodologies and reporting both efficacy and safety (eg, steroid adverse effects) outcomes would provide clearer guidance on tapering rates and how long these patients should ultimately be on corticosteroids. To the knowledge of the authors, there are no such studies underway at this time.
Corticosteroids have been used for rheumatic conditions for almost 70 years, but there is no high level evidence to guide tapering until discontinuation after extended courses of medium- to high-dose treatment regimens. Recommendations for tapering based on largely on expert opinion are available in systematic reviews and clinical practice guidelines for polymyalgia rheumatic and GCA, conditions where corticosteroids are first-line monotherapy.
1. Burns CM. The history of cortisone: discovery and development. Rheum Dis Clin North Am
2. Stahn C, Buttgereit F. Genomic and nongenomic effects of glucocorticoids. Nat Clin Pract Rheumatol
3. Liu D, Ahmet A, Ward L, et al. A practical guide to the monitoring and management of the complications of systemic corticosteroid therapy. Allergy Asthma Clin Immunol
4. Buttgereit F, da Silva JA, Burmester GR, et al. Standardised nomenclature for glucocorticoid dosages and glucocorticoid treatment regimens: current questions and tentative answers in rheumatology. Ann Rheum Dis
5. Moher D, Liberati A, Tetzlaff J, et al. Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement. PLoS Med
6. Higgins JP, Altman DG, Gotzsche PC, et al. The Cochrane Collaboration's tool for assessing risk of bias in randomised trials. BMJ
7. Raine C, Stapleton PP, Merinopoulos D, et al. A 26-week feasibility study comparing the efficacy and safety of modified-release prednisone with immediate-release prednisolone in newly diagnosed cases of giant cell arteritis
. Int J Rheum Dis
8. Viapiana O, Gatti D, Troplini S, et al. Prednisone compared to methylprednisolone in the polymyalgia rheumatica
treatment. Rheumatol Int
9. Russell RW. Giant cell arteritis
. Quart J Med
10. Graham E, Holland A, Avery A, et al. Prognosis of giant cell arteritis
11. Kyle V, Hazleman BL. Treatment of polymyalgia rheumatica
and giant cell arteritis
. I Steroid regimens in the first two months. Ann Rheum Dis
12. Kyle V, Hazleman BL. Treatment of polymyalgia rheumatica
and giant cell arteritis
. II Relation between steroid dose and steroid side effects. Ann Rheum Dis
13. Buttgereit F, Dejaco C, Matteson EL, et al. Polymyalgia rheumatica
and giant cell arteritis
: a systematic review. JAMA
14. Dejaco C, Singh YP, Perel P, et al. 2015 recommendations for the management of polymyalgia rheumatica
. Arthritis Rheum
15. Dasgupta B, Borg FA, Hassan N, et al. BSH and BHPR guidelines for the management of polymyalgia rheumatica
16. Dasgupta B, Borg FA, Hassan N, et al. BSH and BHPR guidelines for the management of giant cell arteritis
17. Mukhtyar C, Guillevin L, Cid MC, et al. EULAR recommendations for the management of large vessel vasculitis. Ann Rheum Dis
18. Hunder GG, Sheps SG, Allen GL, et al. Daily and alternate-day corticosteroid regimens in treatment of giant cell arteritis
: comparison in a prospective study. Ann Intern Med
19. Hahn BH, McMahon MA, Wilkinson A, et al. American College of Rheumatology guidelines for screening, treatment, and management of lupus nephritis. Arthritis Care Res
20. Singh JA, Saag KG, Bridges SL, et al. 2015 American College of Rheumatology guideline for the treatment of rheumatoid arthritis. Arthritis Care Res
21. Ward MM, Deodhar A, Akl EA, et al. American College of Rheumatology/Spondylitis Association of America/Spondyloarthritis Research and Treatment Network 2015 recommendations for the treatment of ankylosing spondylitis and nonradiographic axial spondyloarthritis. Arthritis Rheum
22. Stone JH, Tuckwell K, Dimonaco S, et al. Efficacy and safety of tocilizumab in patients with giant cell arteritis
: primary and secondary outcomes from a phase 3, randomized, double-blind, placebo-controlled trial [abstract]. Arthritis Rheum
. 2016;68(suppl 10): Available at: http://acrabstracts.org/abstract/efficacy-and-safety-of-tocilizumab-in-patients-with-giant-cell-arteritis-primary-and-secondary-outcomes-from-a-phase-3-randomized-double-blind-placebo-controlled-trial/
. Accessed December 14, 2016.
23. Volkmann ER, Rezai S, Tarp S. We still don't know how to taper
glucocorticoids in rheumatoid arthritis, and we can do better. J Rheumatol
24. Choy EH, Smith CM, Farewell V, et al. Factorial randomised controlled trial of glucocorticoids and combination disease modifying drugs in early rheumatoid arthritis. Ann Rheum Dis
25. Boers M, Verhoeven AC, Markusse HM, et al. Randomised comparison of combined step-down prednisolone, methotrexate and sulphasalazine with sulphasalazine alone in early rheumatoid arthritis. Lancet
26. Tengstrand B, Larsson E, Klareskog L, et al. Randomised withdrawal of long-term prednisolone treatment in rheumatoid arthritis: effects on inflammation and bone mineral density. Scand J Rheumatol