EDITORIAL - Corticosteroids in the management of early and established rheumatoid disease

[Guest guest]

Rheumatology Advance Access originally published online on July 31, 2006

Rheumatology 2006 45(9):1058-1061; doi:10.1093/rheumatology/kel230

--------------------------------------------------------------------------

EDITORIAL

Corticosteroids in the management of early and established rheumatoid

disease

E. on1, and H. A. Capell2

1Department of Rheumatology, Southern General Hospital and 2Centre for

Rheumatic Diseases, Glasgow Royal Infirmary, Scotland, UK

Corticosteroid use in rheumatoid arthritis (RA) remains something of a

dichotomy to most rheumatologists. How do we reconcile the promise of

disease-modifying activity in early disease with the possibility of

significant and potentially long-lasting adverse effects? Is the benefit of

the former sufficiently marked to justify the latter? In addition, since the

debate concerning corticosteroid use in RA was re-ignited some years ago,

ideas about the management of RA have moved on. More intense use of

disease-modifying anti-rheumatic drugs (DMARDs), either singly or in

combination, and the proliferation of anti-tumour necrosis factor (TNF-)

agents have had an enormous impact on how we now manage RA [1, 2]. Where do

corticosteroids fit into our current treatment strategies?

RA has also emerged as an independent risk factor for atherosclerosis over

and above those more traditionally recognised (e.g. hypertension, diabetes

mellitus, weight and lipid profile) as a link between systemic inflammation

and accelerated vascular disease has been demonstrated, both in the general

population and in RA [3-5]. Corticosteroids are likely to improve the

inflammatory profile but will this be enough to offset their negative effect

on traditional risk factors?

Patient choice and opinion, central to the day-to-day management of their

disease, is also of prime importance and is a factor often overlooked in the

corticosteroid debate [6].

Many of these issues remain difficult to answer but we attempt here to

revisit [7, 8] and review the status of corticosteroids in the management of

RA in the light of current published evidence.

Corticosteroid use in early RA

The effect of corticosteroids, on both clinical and radiographic parameters,

in early RA has been the subject of considerable research effort in recent

years [9-14].

However, differences in methodology, variability in the type and dose of

DMARD used, differences in corticosteroid dose (5-60 mg) and variable

duration (28 weeks-2 yrs) of therapy combine to make it difficult to compare

the studies directly and to draw firm conclusions regarding efficacy and

toxicity of corticosteroids in early RA.

Efficacy

In a randomized double-blind controlled trial over 2 yrs, Kirwan and the

Arthritis and Rheumatism Council Low-Dose Corticosteroid Group [9] reported

that progression of erosive change was reduced in early RA (<2 yrs) in those

treated with prednisolone (7.5 mg/day) compared with the placebo group. The

study, however, was uncontrolled for DMARD use and a variety of agents were

prescribed by the supervising clinicians [intra-muscular (IM) gold, 8%;

penicillamine, 30%; sulphasalazine, 26%; methotrexate, 4%; others, 3%],

which may have had a bearing on subsequent X-ray progression. In addition,

the two study groups were not particularly well-matched, with the placebo

group containing a higher proportion of erosive patients at the outset.

Initial clinical benefit was not maintained into the second year of the

study and joint destruction resumed when prednisolone was withdrawn [15].

Boers et al. [10] have compared the effect of combination step-down

prednisolone (initially high dose and then tapered and stopped at 28 weeks),

methotrexate (stopped after 40 weeks) and sulphasalazine with sulphasalazine

alone in early RA (<2 yrs) over 56 weeks. Again initial impressive clinical

benefit was seen in the prednisolone group, but was not significant after

the withdrawal of prednisolone. Analysis of the two groups at 28, 56 and 80

weeks showed that radiological progression was reduced in the combination

therapy group. The contribution of methotrexate to reducing X-ray

progression may have been significant in this study, but unfortunately a

methotrexate/sulphasalazine arm was not included with the result that the

true effect of prednisolone is difficult to extrapolate.

In the van Everdingen et al. [11] study of low-dose prednisolone vs placebo

in early RA, 40 patients received placebo and 41 received 10 mg of

prednisolone. After 6 months, sulphasalazine could be used as rescue-this

design does not reflect current rheumatological practice. While radiographic

improvement favoured the steroid group, adverse events including weight

gain, hyperglycaemia and vertebral fractures were a problem in this group.

Capell et al. [12], reporting on behalf of the West of Scotland Early

Rheumatoid Arthritis Corticosteroid Therpy (WOSERACT) study group, evaluated

the effect of low-dose prednisolone (7 mg/day) compared to placebo on

disease activity and X-ray progression in early RA (<3 yrs) in a randomized

controlled trial (RCT) over 2 yrs. The study was controlled for DMARD as all

patients received sulphasalazine at the outset, with the option of an

alternative DMARD later if necessary. The two treatment groups were

well-matched for all parameters at baseline due to the technique of

minimization [16, 17] in the allocation process. No significant differences

in radiological scores, clinical indices or laboratory measurements were

observed at baseline or 2 yrs.

In a more recent double-blind, placebo-controlled trial over 2 yrs,

Wassenberg et al. [13] reporting on behalf of the Low-Dose Prednisolone

Therapy Study Group, compared the effect of prednisolone (5 mg/day) with

placebo in early RA (<2 yrs) patients who were also started on DMARD

therapy, gold sodium thiomalate or methotrexate, at baseline. There was no

significant difference in the number of patients on each DMARD between the

prednisolone and placebo groups. Radiographic progression was noted to be

significantly less with prednisolone than placebo, and interestingly the

greatest difference in progression rate was noted in the first 6 months of

treatment. However, scrutiny of the baseline characteristics reveals a

tendency (not reaching significance) towards more erosive disease in the

placebo group, 76.7% compared with 71.3% in the prednisolone group. In

addition, >50% of the patients were given gold as their primary DMARD whilst

<40% received methotrexate which does not reflect most rheumatologists'

current practice.

Svensson et al. [14] assessed the efficacy of prednisolone (7.5 mg/day) vs

placebo in early RA (1 yr) in a 2 yr randomized trial. The choice of DMARD

was left to the treating physician. In the prednisolone group, 50% started

methotrexate and 35% received sulphasalazine, the corresponding values for

the placebo group were 53 and 37%. The study group observed that

radiographic progression of erosions was significantly less in the

prednisolone-treated group. However, it is not clear if the study was

adequately powered as the authors admit that no formal sample size

calculation was performed. In addition, 43 patients were excluded from the

study because they were considered to need corticosteroid therapy due to

highly active disease. Thus, we have no information about a group likely to

be highly erosive. Due to these methodological difficulties, the results

obtained are difficult to interpret.

In the five recent low-dose prednisolone studies outlined above, there are

methodological differences, variation in corticosteroid dose and duration,

no uniformity regarding concomitant DMARD therapy and different X-ray

scoring techniques used to assess progression of erosions. The definition of

early RA itself is wide at between <1-3 yrs. Herein lies the challenge for

the clinician. A pragmatic assessment may be that low-dose prednisolone is

efficacious in the short-term in early RA, clinically and radiologically,

but as yet there is no convincing evidence of sustained benefit which would

radically alter routine practice.

Toxicity

The side-effect profile of corticosteroids has been well-documented since

their earliest use in RA [18, 19]. Physicians are familiar with the

characteristic features of weight gain and redistribution, skin thinning,

osteoporosis, diabetes, hypertension, cataracts and mood disturbance, albeit

at doses we would now consider unsuitable in the routine management of RA.

However, little detailed evidence is available regarding the side-effect

profile of low dose corticosteroids in early disease. Da Silva et al. [20]

in their review of the safety of low-dose corticosteroid treatment in early

RA conclude that safety data from recent trials suggests that adverse

effects are modest and may not be significantly different from placebo.

Others are less convinced [21, 22].

In fact, scrutiny of recent RCTs [9-14] reveals that adverse events have

been variously monitored and reported. Significant weight gain was reported

by Boers et al. [10], Capell et al. [12], Wassenberg et al. [13] but not by

Kirwan et al. [9] or Svensson et al. [14] in the prednisolone-treated

groups. Wassenberg et al. [13] noted a mean weight increase of 5 kg in the

prednisolone group compared with a mean increase of only 0.3 kg in the

placebo group, a finding likely to be deeply unpopular with patients,

especially women. Blood pressure measurements remained stable as did bone

density in those centers where dual energy X-ray absorptiometry was

available. Blood glucose, when measured remained largely unaffected. Lipid

profile was observed in only one study [12] and was stable over 2 yrs in

those patients in whom it was recorded. Wassenberg et al. [13] assessed for

cataract and glaucoma at the beginning and end of their study. Over the 2

yrs, three patients in the prednisolone group and none in the placebo group

developed glaucoma. None of these trials systematically monitored effects on

the skin. Whilst these are usually not considered serious by physicians, the

reality of thin, fragile skin can be considerable morbidity for the patient.

Such skin is prone to bruising, tearing and poor healing especially on the

exposed areas of the forearm and shin.

The recent RCTs were powered to establish the efficacy of low-dose

corticosteroids in early RA and the number of patients included reflect

this. Most of the major studies [9, 10, 12, 13] include <100 patients in

either their prednisolone or placebo arms and none of these have

comprehensively documented weight, blood pressure, blood glucose, lipid

profile, bone density/fractures, infection rate, skin effects, etc. in all

the patients. Much larger numbers of patients and a more detailed search for

adverse events would be required to adequately assess toxicity. An analysis

of non-responders, clinical and/or radiological, would also be valuable. In

addition, some individuals may be more susceptible to particular adverse

effects than others; age, gender and comorbidity might be relevant here. We

should not, therefore, be complacent about the safety of low-dose

corticosteroids in early RA.

Corticosteroid use in established RA

Corticosteroids are widely used for short/medium-term symptom relief during

disease flares or induction of DMARD therapy in patients with established

RA. Several studies using a variety of agents [e.g. oral prednisolone, IM or

intravenous (IV) methylprednisolone] in combination with DMARDs support this

practice [23-26] but conclude that the beneficial effects (clinical and

radiological) are short-lived. One 6 month RCT of sulphasalazine combined

with pulses of methylprednisolone or placebo showed no differences between

the two groups [27]. In addition, Saag et al. [28] reported in a historical

cohort of RA patients that low-dose (<=5 mg/day) long-term prednisolone is

associated with the development of adverse events in a dose-dependent

fashion; although, they noted that disease severity was an important

confounding factor.

Choy et al. and the Intramuscular Methylprednisolone Study Group [29]

conducted a 2 yr RCT to establish the benefits of IM depomedrone vs placebo

in patients with established RA (mean disease duration 16 yrs) whose disease

was inadequately controlled by their existing DMARDs. After screening and

consent, 91 patients were followed up. The steroid group (n = 48) received

120 mg IM depomedrone monthly (equivalent to 5 mg prednisolone/day) whilst

the placebo group (n = 43) received IM sterile normal saline. Improvement in

disease activity was better initially in the steroid-treated group but by 6

months no difference remained. A small but significant reduction in

progression of erosive change was noted in the steroid group. However, more

adverse events occurred in those patients receiving steroids compared with

placebo (55 compared with 42) and this was especially marked in those

effects particularly associated with corticosteroid use (16 compared with

two). The following were noted in the steroid group: hypertension (n = 4),

facial swelling (n = 3), bruising (n = 3), osteoporosis (n = 2, one patient

with vertebral fracture), diabetes mellitus (n = 1), myocardial infarction

(n = 1), hypercholesterolaemia (n = 1) and iatrogenic 's disease (n =

1). In contrast, only two placebo-treated patients had similar effects:

hypertension (n = 1) and weight gain (n = 1).

The authors conclude, in the light of the significant side-effect profile,

that despite the initial benefits of IM depomedrone, RA patients should not

receive long-term steroids in addition to their DMARD(s) when their disease

is sub-optimally controlled but should instead be treated with additional or

alternative DMARDs. This view is supported by Durez et al. [30] who

demonstrated in their short-term (6 weeks) randomized comparative study of

IV pulse methylprednisolone vs infliximab in RA patients (median disease

duration 10 yrs) with active disease despite methotrexate that TNF blockade

is superior to pulsed steroid therapy.

Does route of administration of corticosteroids matter?

Furtado and colleagues [31] studied 69 RA patients with 6-12 swollen joints

and randomized them to polyarticular injection (6-8) with triamcinolone or

equivalent doses of IM triamcinolone. Over 6 months, the intra-articular

(IA) group had significantly better American College of Rheumatology 20, 50

and 70 responses, and fewer adverse events than IM group. Blood pressure was

lower in the IA group and they sought help less often as measured by calls

to the physician or hospital unit.

Corticosteroids and cardiovascular risk

Since the inception of most of the recent RCTs of low-dose prednisolone in

the treatment of early RA, evidence revealing an association between

accelerated atherosclerosis and RA has been accumulating and has been

recently reviewed [5]. In addition to traditional risk factors such as

hypertension, lipid profile, weight, etc. systemic inflammation is likely to

play an important part in the aetiology of atherosclerosis. Conventionally,

corticosteroids have been thought to exacerbate cardiovascular morbidity

through negative effects on traditional risk factors, but the true direction

of their effect is likely to be a more complex interaction, perhaps offset

by their beneficial effects on inflammatory markers [32]. More work is

needed before we can be sure. Clarification is particularly important for

those of us who treat populations already at high cardiovascular risk [33].

Summary

The use of corticosteroids in RA is clearly something of a clinical

balancing act, the key to which would seem to be judicious timing. Early

initiation of DMARD therapy in new onset RA, with escalation as required to

achieve disease control, is the essence of current good practice.

Corticosteroids used early and for short periods, either orally or

parentally, are an effective adjunctive measure. The caveat being that

detailed information about their side-effect profile in this setting is

lacking. In established RA, however, the evidence suggests that

corticosteroids should be reserved for short-term use during flares of

disease activity or as bridge therapy until the efficacy of a DMARD is

established. Longer use cannot be justified as the clinical and radiological

benefits are relatively small and adverse events not inconsequential.

Despite some continuing uncertainties and reservations, the results of

recent RCTs have undoubtedly refined our use of, and subtly altered the

position of, corticosteroids in the management of both early and established

RA. They remain, therefore, one of our most powerful and useful therapeutic

interventions and it is our responsibility to use them wisely.

The authors have declared no conflicts of interest.

http://rheumatology.oxfordjournals.org/cgi/content/full/45/9/1058

Not an MD

I'll tell you where to go!

Mayo Clinic in Rochester

http://www.mayoclinic.org/rochester

s Hopkins Medicine

http://www.hopkinsmedicine.org