Corticosteroids: the knee jerk response

Once upon a time there were two guidelines mentored to rheumatology fellows on the first day they reported to work: a) there’s no short-term solution for a long-term problem, and b) do no harm. Then along came familiarity with corticosteroids (CSs), coupled with the hero worship of any physician who could make miraculous things happen virtually overnight. CSs became routine treatment for polymyalgia rheumatica (PMR), systemic lupus erythematosus (SLE), vasculitis, and rheumatoid arthritis (RA).

How many of you reading this column have PMR patients who cannot discontinue CSs because of prompt recurrence of their inflammation (IM)? It’s unfortunate if they have developed compression fractures, tissue paper skin, diabetes, hypertension, or other side effects from conventional use of CSs in this condition, because that is the “proper” treatment¹. Well, as it turns out, despite decades of perpetuation of this advice, CSs may not be the proper initial treatment.

A recent published study examining the natural course of PMR without the use of CSs strongly suggests that traditional concepts regarding this syndrome are in need of reassessment, including the “diagnostic trial” of CSs². In this study, 77% of acute PMR patients evolved over time into a definitive diagnosis of seronegative RA, which in turn was dramatically responsive to treatment with hydroxychloroquine (HCQ). Once educated with alternative choices, this study group elected to avoid the potential side effects of CSs and wait for HCQ to start to work.

Which scenario is worse: (a) the initial time spent explaining the reality of an illness and its treatment, or (b) the subsequent time spent during innumerable follow-up visits managing all the treatment-induced side effects of CSs?

Terminology and immunology make strange bedfellows. We freely substitute lupus cerebritis for central nervous system (CNS) involvement in SLE, and then justify it by demonstrating the presence of antibodies directed against varied constituents of neuronal tissues. Pathologists, however, are often at a loss to demonstrate any actual brain IM in damaged areas of the CNS. This is directly analogous to the problems exemplified in patients with autoimmune retinopathy, where pathologically there is an absence of IM in deteriorating retinal components despite the serological presence of anti-retinal antibodies. Perhaps in these conditions the presence of antibodies is a reflection of damaged tissues rather than the cause (i.e., the antibodies are epiphenomena). Many authors have demonstrated that the release of organelles from traumatized cells can activate a host of inflammatory pathways and also lead to the production of autoantibodies³. Even the relatively new syndrome, ASIA (autoinflammatory syndrome induced by adjuvants), with its multiple causes and outcomes, admixes terminology and immunology in confusing ways. How are chronic CSs going to favorably affect any of the above conditions?

Thirty-five years ago I was reviewing an open lung biopsy on a patient with Wegener’s granulomatosis (the old terminology). The literature from the 1970s demonstrates that once the disease fulminates the median survival with CSs alone is five months, and roughly 80% of patients expire within one year on CSs in any dose. In the era prior to rituximab trials I asked myself: “why would I want to treat with both corticosteroids and cyclophosphamide; why shouldn’t I use oral cyclophosphamide alone?” In other words, if medicine A doesn’t work, why should it be mixed with medicine B – why not give medicine B by itself? And that is exactly what I have done in 18 similar patients over the past 35 years where my advice was solicited before any CSs were administered. Results have been superb without the usual morbidity and mortality accompanying concomitant CS use, and I have published a concise report on several of these cases⁴. Rituximab trials in necrotizing vasculitis need to consider this approach. The rationale for all of this is augmented by knowledge available since 1979 that patients taking CSs make too many clots and in turn don’t lyse them well⁴. In light of this consider polyarteritis nodosa, where tissue ischemia is augmented by the secondary phenomenon of thrombus formation at the sites of inflammatory blood vessel damage. The use of CSs in this disorder may actually precipitate an acceleration of the patient’s life threatening course despite their well known anti-inflammatory effects.

In 1989, at the annual meeting of the American College of Rheumatology, a prominent rheumatologist (whom I shall leave nameless) got up in front of thousands of attendees and flatly stated that every rheumatoid arthritis patient should be placed on a modest dose of CSs indefinitely. Prior studies stating that the radiographic progression of RA was unaltered by chronic CS use were balanced by subsequent studies asserting just the opposite. Other investigators have examined the long term use of 3 mg daily prednisone in RA and have reported conflicting results ranging from those of exceptional benefits to those of intolerable side effects^5,6.

Leaving aside the options for utilizing intermittent short courses and/or moderate term courses of CSs in RA, with all the armamentarium that we now have for RA how does one justify prescribing chronic uninterrupted daily CSs indefinitely for routine care in the average case?

We are all fortunate to have CS treatment available for conditions such as giant-cell arteritis (GCA), acute optic neuritis, transverse myelitis, acute hypersensitivity disorders, Loffgren’s syndrome, severe asthma, inflammatory bowel disease, and the hospitalized gouty patient, to name a few. And with regard to GCA, no steroid sparing agonists have been definitively identified.

Nevertheless, patient education, therapeutic salicylates, and patient preferences would be wiser first steps in treating this patient population. Why? As far back as 1971 we knew that only 25% of patients with GCA manifest any transient visual disturbances, and only two percent of those go suddenly blind without any warning whatsoever⁷. In addition, it has now been shown that fever, night sweats, and weight loss are not discriminating features that help differentiate GCA from either pure PMR or late onset seronegative RA². Plus, we also know that it has been standard methodology for decades not to perform a temporal artery biopsy in patients with PMR who have no symptoms above the neck such as jaw pain, tongue pain, headaches, etc.

As rheumatologists we are expected to render cost effective evidenced-based care that will alleviate misery, not cause it. Are we so hypocritical that we can demonize a colleague on Monday for having subjected a patient to the ravages of chronic CS use, and then reverse course on Tuesday by prescribing identical treatment for another patient in order to expedite care on a hectic day? Leave the knee jerk response to the neurologists. Patients need reliable information on CSs to become their own advocates, and we should not emulate the legions of cell phone texters who worship the flight from conversation. Indeed, the routine use of CS use is a conversation that is in need of reassessment.

Transparency

Declaration of funding

There has been no financial support, nor any other benefits from any commercial source, for the work reported in this manuscript.

Declaration of financial/other relationships

A.E.B. has disclosed that he has no significant relationships with or financial interests in any commercial companies related to this study or article.

CMRO peer reviewers on this manuscript have no relevant financial or other relationships to disclose.