CNS Involvement in Rheumatoid Arthritis: An Expert Perspective

Central nervous system (CNS) involvement in the disease course of rheumatoid arthritis (RA) is relatively uncommon. However, CNS disorders occasionally occur in patients with RA as manifestations of the disease process, unrelated comorbidities, or adverse effects of biological agents used to treat RA.

Fabiola Atzeni MD, PhD, from the University of Messina, Italy, and colleagues recently performed a review of literature published on the topic of CNS involvement in patients with RA and the incidence of new-onset CNS diseases in patients with RA receiving antitumor necrosis factor (anti-TNF) or non-anti-TNF biologics. Seventy-six articles published between 1989 and 2017 were retrieved from the Medline database and were analyzed in the review. Results showed that:

• RA directly affects the CNS, giving rise to aseptic meningitis, vasculitis, and rheumatoid nodules within the brain and meninges.
• CNS comorbidities including vascular events, infections, neurodegenerative diseases, and demyelinating syndromes can affect patients with RA. Although not directly related to RA, these conditions may be influenced by prolonged systemic inflammation and therapies that block various inflammatory pathways.
• CNS diseases related to the use of anti-TNF drugs (infliximab, adalimumab, etanercept, certolizumab, golimumab, and their respective biosimilars) include demyelinating diseases and opportunistic infections of the brain and meninges. However, the CNS effects of anti-TNF therapy on patients with RA do not appear to be wholly detrimental. Systemic anti-inflammatory properties of anti-TNF therapy may be protective against dementia-related neurodegenerative disorders.
• CNS diseases related to the use of non-anti-TNF biologics in RA have not been well characterized as a result of relatively small cohorts of patients and short periods of observation. Progressive multifocal leukoencephalopathy, a demyelinating disease of the CNS sustained by polyomavirus JC infection, may occur very rarely in rituximab-treated patients. Adverse CNS effects of anakinra, abatacept, and tocilizumab include headache.

To glean a greater understanding of CNS involvement in RA and its clinical implications, Rheumatology Advisor asked Elena Schiopu, MD, associate professor at the University of Michigan Medical School, Ann Arbor, to comment on Dr Atzeni’s review:

“CNS involvement, while described [in the medical literature], does not come to the forefront of clinicians’ assessment among the extra-articular manifestations of RA. As rheumatologists, unless also trained in geriatric medicine, we don’t routinely assess our patients’ cognitive function through tools like the Mini-Mental State Exam or the Montreal Cognitive Assessment, but we definitely are aware of the potential side effects of the TNF blockers, including the hard stop of demyelinating disease, but also opportunistic CNS infections, and the potential complications of poorly treated RA, such as rheumatoid nodules or lepto-meningitis. We also worry about the occurrence of [progressive multifocal leukoencephalopathy] in patients treated with rituximab, but also any other immunosuppressant, even though there are no screening tools to mitigate this terrible disease.

The Atzeni article is the first to evaluate the published data pertinent to any CNS involvement in the setting of RA, whether as part of its clinical manifestations or side effects of the drugs used to treat it. As expected, there was not an abundance of articles (76 papers over the course of almost 30 years), most of them case reports or series, and baseline and longitudinal data was missing.

While CNS vasculitis is rare, CNS vascular events are more common in RA compared to the general population, likely related to the increased inflammatory cytokines but also due to the propensity to clot. And of course, we use glucocorticoids and [nonsteroidal anti-inflammatory drugs] to treat inflammation and pain in RA, and both of these drugs are associated with an increased incidence of strokes. Comorbid [antineutrophil cytoplasmic antibody-associated] vasculitis, or primary Sjögren’s, or even lupus are obviously red flags to point our attention toward CNS involvement. Dementia is also a subject of much attention, as it is becoming the leading cause of mortality in certain countries, and it has no cure. This article found 2 papers describing an increased relative risk and odds ratio for dementia and Alzheimer’s in patients with RA when compared to the general population, with inflammation and disease duration being positively associated. Depression, anxiety, and mood disorders are also prevalent in RA, but comparable to other chronic disorders.

As rheumatologists at the forefront of manipulating the immune function, we are walking a very fine line as we balance chronic disease burden against the risk for opportunistic infections and malignancies. By treating patients for a long time, we do have a unique insight into patients’ cognition and mood, as we connect with the patients’ families, who serve as surrogate markers of both of these functions. We have now grown used to the systemic manifestations of autoimmune disorders. The CNS, even though carefully isolated by the blood-brain barrier, could be affected in many ways. While it is easier to recognize seizures or focal deficits due to infections or rheumatoid nodules, dementia or mood disorders tend to go unnoticed. We need to be particularly observant in clinic, and include a baseline brief cognitive/mood assessment in our evaluation, as CNS function as evidenced by mood and cognition could change over time. It is difficult to outline recommendations to clinicians based on this relatively small number of published papers, but this article is timely and a good reminder that CNS complications still exist.”

Reference

Atzeni F, Talotta R, Masala IF, Gerardi MC, Casale R, Sarzi-Puttini P. Central nervous system involvement in rheumatoid arthritis patients and the potential implications of using biological agents. Best Pract Res Clin Rheumatol. 2018;32:500-510.