Round 1 : Palindromic Rheumatism

By Gordon Lam, MD

Fellow, Division of Rheumatology
Johns Hopkins University School of Medicine

Release Date: December 1, 2005
Expiration Date: December 1, 2007

Dr. Lam has no significant financial interest or relationships to disclose.

The term palindromic rheumatism, our topic of discussion today, is distinct from a palindrome, which is a phrase that is symmetric about a central axis and hence reads the same forward as backward. Rather, palindromic rheumatism is an idiopathic, periodic arthritis characterized by multiple, transient, recurring episodes of mono- or oligo-arthritis associated with tissue swelling around the involved joints. Episodes last for a few hours or a few days and then spontaneously resolve. Between episodes, there are no residual effects. The descriptor “palindromic” is used based on its Greek root palindromos, which means “to come and to come again,” which aptly describes this phenomenon.

Although first described more than 60 years ago, not much is known about palindromic rheumatism. It is an extremely rare condition, and often, doctors who don’t recognize it treat it as rheumatoid arthritis. Today’s Rounds will hopefully help to shed some light on it by discussing:

  • The epidemiology of palindromic rheumatism;
  • The clinical, pathological, and laboratory aspects of the disease;
  • Its genetics;
  • Prognosis and treatment; and
  • Its progression to other more chronic forms of diseases, specifically rheumatoid arthritis.

In reviewing these topics, we also will touch on two controversial issues of this disease, that is, the mechanisms of its pathophysiology; and its classification, or nosology.

First let’s examine the case of a 48-year-old woman who came to our clinic complaining of joint pain, which had begun about 2½ years prior to her presentation. The joint pain first presented as sudden, acute, intense pain localized to her right shoulder. The pain was rated 7 out of 10 in intensity on the analog pain scale, and it was described it as a burning, searing, and sharp sensation. Swelling, redness and exquisite tenderness to touch were associated with the joint pain. At that initial onset, it involved only her right shoulder and the structures immediately around it. The pain quickly evolved, and within a few hours it had reached its maximum intensity. Interestingly, the patient had no preceding morning stiffness or other constitutional symptoms. Movement exacerbated the pain. She tried Tylenol and immobilization of her right arm, but these did not help. Her primary care physician recommended ibuprofen, which also provided no relief. Then, just as quickly as the pain came on, it spontaneously resolved. Our patient suffered no subsequent complications or consequences.

However, a month later, the pain returned in a very similar fashion. It came on very quickly and again only involved one joint. It evolved over the course of a few hours and was excruciatingly intense. This time, though, instead of affecting her right shoulder, it was her left wrist. After 24 hours, the pain again mysteriously abated. Over the course of the next year, our patient had these attacks about once a month. They affected her fingers, wrists, elbows, shoulders, hips, knees, and even her jaw at one point, with each episode only attacking one joint. Throughout this, she had no constitutional symptoms such as fevers, fatigue, or unintentional weight loss. Between each of these episodes, she was completely asymptomatic, and her activities of daily living were never compromised.

Our patient had no history of Lyme disease or sexually transmitted diseases. Laboratory studies drawn by her primary care physician were unremarkable: her sedimentation rate was 11mm/hr; C-reactive protein was 0.9 mg/dL; rheumatoid factor was negative on two occasions; ANA was negative; Lyme serologies were negative, and all cultures, including those for gonorrhea and Chlamydia, were negative. The patient had no significant past medical history, and her social history and family history were non-contributory. Despite her episodic arthritis, she never missed a day of work and was still able to enjoy her hobbies. She did not take any prescription or over-the-counter medications.

At our first clinic visit with the patient, our physical examination was unremarkable. There was no evidence of synovitis. She had no restrictions of movement, effusions overlying any of her joints, or evidence of any structural defects. Skin examination revealed no evidence of nodules, rashes, or tophi. Laboratory evaluation was completely unremarkable, with the exception of her level of anti-cyclic citrullinated peptide (anti-CCP) antibodies, which was markedly elevated at over 100 units. A repeat rheumatoid factor was negative.

Thus, in summary, our patient was a 48-year-old female with no relevant family history who presented with reports of an episodic, migratory mono-arthritis for 2½ years that was not associated with any constitutional symptoms, morning stiffness, or physical examination findings during disease-free intervals. The only laboratory abnormality was a markedly elevated anti-CCP.

A phenomenon similar to what our patient experienced was first described in 1944 by Edward Rosenberg and Philip Hench, who incidentally was awarded the Nobel Prize in medicine in 1954 for his discovery of cortisone. Rosenberg saw his first patient with palindromic rheumatism in 1928. In 1944, they published a paper in the Archives of Internal Medicine, entitled “Palindromic Rheumatism,” which coined the phrase that is still used today. They reported 34 cases of this phenomenon, collected between 1928 and 1939 and all from the Mayo Clinic. After describing each case in great clinical detail with follow-up data of many years, they concluded that palindromic rheumatism represented a new, distinct clinical entity.

However, it wasn’t until approximately 15 years later that this condition was revisited by Ansell and Bywaters, who published a case series on 28 patients with palindromic rheumatism. They noted that the majority of their patients eventually developed rheumatoid arthritis. So, in stark contrast to Hench and Rosenberg, who postulated that this was a distinct immunologic entity, Ansel and Bywaters concluded that palindromic rheumatism was merely a variant of rheumatoid arthritis. This started a controversy as to exactly what palindromic rheumatism is and what its pathophysiology represents. Since then, a multitude of papers over the next 18 years by Mattingly, Williams, Wajed, Bregeon, and Hannonen reported various patients with syndromes similar to that originally described by Rosenberg and Hench. In these reports, a third to a half of all cases eventually evolved into rheumatoid arthritis. Yet, none of them found a clinical, immunologic, or genetic factor that could reliably predict who would develop rheumatoid arthritis and who would maintain a course of palindromic rheumatism.

In 1986, Pasero and Barbieri proposed five criteria required for diagnosis of palindromic rheumatism:

  1. Six-month history of brief, sudden onset and recurrent episodes of mono-arthritis or polyarthritis;
  2. Direct observation by a physician of at least one attack;
  3. Involvement of three or more joints, although they could be involved at different times;
  4. Complete absence of radiographic findings; and
  5. Exclusion of all other arthritides.

This set of criteria was never adopted by the American Rheumatism Association or by the American College of Rheumatology, but it did add fuel to the controversy of palindromic rheumatism. By proposing diagnostic criteria, Pasero and Barbieri imposed that this, in fact, represented a distinct clinical entity, separate from rheumatoid arthritis.

The difficulty in characterizing palindromic rheumatism is simply that this is an extremely rare disease. Depending on which report you read, it has a prevalence of approximately 1/8 to 1/20 that of rheumatoid arthritis. Unlike rheumatoid arthritis, which affects mostly women, men are equally affected by palindromic rheumatism, and their age range can span from the 20s to the 70s. There is no known ethnic predisposition to the disease. Clinical features have very sudden onset and progressive intensity, which usually peaks within a few hours. Attacks can last from two hours to more than two weeks, but most episodes last two days on average. Constitutional symptoms and morning stiffness are rare. Different joints can be affected with each attack, and any joint in the body is susceptible, although the hands, particularly the metacarpophalangeal and the proximal interphalangeal joints, are the most commonly involved. The spine and the jaw are rarely affected. Interval periods are symptom free.

Joint Mean % of Patients Range of % of patients
MCP/PIP 91 74-100
Wrists 78 54-82
Knees 64 41-94
Shoulders 65 33-75
Ankles 50 10-67
Feet 43 15-73
Elbows 38 13-60
Hips 17 0-40

Palindromic rheumatism also has associated cutaneous manifestations. Although subcutaneous nodules may form, they are different from those commonly seen in rheumatoid arthritis–they are transient, smaller, found mainly in the hands, and nodules lack any central fibrinoid necrosis or palisading mononuclear cells. A second cutaneous manifestation, seen in the slide below, is neutrophilic dermatitis, characterized by well demarcated, violaceous papules and plaques that are edematous.

neutrophilic dermatitis imageThey are usually symmetric and occur on extensor surfaces. Although rare, they are commonly seen in the hand.

In terms of objective findings, inflammatory markers usually are not elevated during the symptom-free intervals, but they may be elevated during flares of the disease. Radiographs are always normal. Erosions, joint space narrowing, and periarticular osteopenia are absent.

The differential diagnosis of palindromic rheumatism is broad and includes rheumatoid arthritis, crystalline arthropathies, gout, seronegative spondyloarthropathies such as reactive arthritis or inflammatory bowel disease-related arthritis, Whipple’s disease, Behcet’s syndrome, relapsing polychondritis, intermittent hydrarthrosis, and familial Mediterranean fever.

The association of rheumatoid factor and anti-CCP in patients with palindromic rheumatism is perhaps the most fascinating aspect of this condition. Rheumatoid factor is found in 40 – 65% of these patients according to some studies, and its presence has been associated with a more severe form of disease as well as a higher probability of eventually developing classic rheumatoid arthritis.

Reference PR(%) Pr->RA(%) F/U
Hannonen et al. 11/25(44) 29/35(83) 5 yrs
Bregeon et al. 3/16(19) 16/25(64) 10 yrs
Wajed et al. 2/22(9) 16/17(94) 10 yrs

For example, in Hannonen’s study, 83% (29 of 35) of patients with palindromic rheumatism who were rheumatoid factor positive eventually developed rheumatoid arthritis, compared with only 44% of patients (11 of 25) who maintained palindromic rheumatism.

Anti-CCP was examined in only one paper by Salvador (Rheumatology, 2003). In it, anti-CCP was present in 56 percent of 32 patients with palindromic rheumatism; rheumatoid-factor was found in 15 of the 32 (46%). Rheumatoid factor and anti-CCP were highly correlated: in 18 patients who were anti-CCP positive, 10 of them were also rheumatoid factor positive. The authors concluded that because anti-CCP is specific for rheumatoid arthritis and because rheumatoid factor is correlated with rheumatoid arthritis, palindromic rheumatism is, therefore, on the same spectrum of disease. They reasoned that palindromic rheumatism may be considered an abortive form of rheumatoid arthritis.

When Hench and his colleagues first described palindromic rheumatism in patients, they postulated that genetics played a role, though they had no proof of family incidence. To this day, good data on the genetics of palindromic rheumatism is still lacking. At this point, the best conclusion one can draw is that there is an absence of a striking association with HLA-DR4, which suggests that palindromic rheumatism is not simply a variant of rheumatoid arthritis, and perhaps not in the same spectrum of disease.

In terms of treatment, there have been no randomized, controlled trials investigating treatment of palindromic rheumatism. People have tried all of the conventional therapies for rheumatoid arthritis, with reported response rates from 3 – 68% effectiveness. Even anti-malarials have been used, with 15 – 80% response; sulfasalazine appeared effective in half of all trials; gold injections produced responses in 20 – 68%; and colchicines reportedly induced remission in four of five patients. But because the total number of patients treated is so small, no statistical significance can be ascertained from these data. In our patient, we initially prescribed non-steroidal anti-inflammatory drugs for her acute attacks, although they had little, if any, clinical benefit. We discussed using prednisone during acute attacks, but our patient opted to avoid this agent. Surprisingly, there is little in the literature about using corticosteroids for this condition.

With regard to the natural history of palindromic rheumatism, three patterns have emerged:

  1. Clinical remission of attacks;
  2. Recurrent attacks without persistent joint involvement; and
  3. Evolution into a more chronic disease, such as rheumatoid arthritis.

So, in returning to the age-old controversy, the question remains: Is palindromic rheumatism a form of rheumatoid arthritis, or is it a distinct entity?

Points that argue for it being withing same spectrum of disease include these facts:

  • Up to 50 percent of patients with palindromic rheumatism eventually progress to rheumatoid arthritis
  • There is a high prevalence of rheumatoid-factor and anti-CCP in patients with palindromic rheumatism.
  • Cutaneous nodules, which is a manifestation of rheumatoid arthritis, are also present in palindromic rheumatism (although the histopathology is different).
  • There is some response to common treatments of rheumatoid arthritis, such as NSAIDS, anti-malarial agents, gold injections, and sulfasalazine. However, the responses are variable and not as effective as in rheumatoid arthritis.

On the opposite end of the argument, there are equally compelling data that suggests that palindromic rheumatism is a distinct clinical entity:

  • The pattern of inflammation differs, with frequent remissions separated by symptom-free intervals and no apparent accumulation of disease.
  • There is no bone or joint destruction.
  • Constitutional symptoms are absent or rare.
  • There is no strong HLA association.
  • The demographics of palindromic rheumatism differs from that of rheumatoid arthritis, specifically, it affects men and women equally (whereas rheumatoid arthritis favors women).

In conclusion, palindromic rheumatism is an idiopathic, periodic arthritis marked by multiple and recurrent attacks affecting one to a few joints, with tissue inflammation around or adjacent to them. Episodes usually last from a few hours to several days, and then the attack disappears just as peculiarly as its onset. There are no residual effects, such as bony abnormalities or impairments to the patient’s overall functioning status. The etiology of palindromic rheumatism is unknown, similar to that of rheumatoid arthritis, although anti-CCP is associated with it. Approximately 1/3 to a 1/2 of patients eventually evolve to a more classic form of rheumatoid arthritis. Risk factors to this appear to be rheumatoid factor positivity, female sex, and involvement of the hands. There is no genetic association known, and no clinical trials for treatment have been performed.

In the future directions for this field of palindromic rheumatis include: 1) the need for long-term follow-up in larger cohorts of patients; 2) recognition of associations with other diseases; 3) development of a standardized set of diagnostic criteria; and 4) systematic studies of traditional and novel therapies, such as methotrexate, leflunomide, and anti-TNF agents.

 

Updated: March 21, 2012

Arthritis Center

About Arthritis Center

Founded in 1998, the Arthritis Center at Johns Hopkins is dedicated to providing quality education to patients and healthcare providers alike.