Vaccines Linked to Arthritis



Yehuda Shoenfeld Department of Internal Medicine B, Sheba Medical Center Tel Hashomer and Center for Autoimmune Diseases, Sackler School of Medicine, Tel Aviv University

Vaccination and the possible autoimmune side effects have been discussed recently in several articles (1-4). Not omitting for one moment the immeasurable benefits of vaccination, the question of whether this practice may be one of the factors implicated in the onset of autoimmune illness has arisen (3).

Based on existing case reports, there seems to be at least a temporal connection between several vaccines and autoimmune phenomena; however, no causal relation has been proven yet. In this cutting-edge article, we wish to focus our attention specifically on the combination of immunization and arthritis.

The occurrence of arthritis has been described after various types of vaccines (Table 1) have been given and can be divided into isolated or reactive arthritis (poly or monoarticular) and arthritis as part of a systemic autoimmune disease (such as systemic lupus erythematosus [SLE] or rheumatoid arthritis [RA]). Some vaccines have been implicated more often than others. We will review the existing reports, within the last 10 years, on arthritis following vaccination.

Arthritis and Hepatitis B Vaccine

In the last 10 years, at least 32 cases of arthritis following immunization against hepatitis B have been described. Some cases have been isolated inflammation of the joints and others were part of frank RA (Table 1).

In 1990, two cases of arthritis were reported (5,6). One patient developed erythema nodosum and polyarthritis and the other reactive arthritis shortly after receiving an HBV vaccination. More reports were published during the following years (7-13). Some of the patients described had high titers of rheumatoid factor in the serum, even though they did not fulfill other criteria for frank RA (8). Others were carriers of genetic markers predisposing to autoimmune diseases (13). A 1998 report (13) described 11 patients who developed arthritis after receiving HBV recombinant vaccine. Ten of the 11 patients reviewed fulfilled the revised ARC criteria for RA. Nine of those required disease modifying drugs. Five subjects were HLA-DR4 positive. Nine of eleven patients genotyped for HLA-DR and DQ expressed the RA shared motif in their HLA class II genes. Findings raised the possibility that HBV recombinant vaccine may trigger RA in genetically prone individuals. An additional case reported in 1996 (10) supports, to a certain extent, the hypothesis that people genetically prone to develop autoimmune disease will more readily develop arthritis after vaccination (of any kind). In this case, a 44-year old man who had had myasthenia gravis 20 years earlier developed arthritis, hypercalcemia, and lytic bone lesions shortly after administration of HBV recombinant vaccine. It is noteworthy that a definite connection between the arthritis and the vaccine itself has not been established (serologically) in any of the cases mentioned so far. Additional cases of frank RA where the HBV vaccine was the trigger for the onset of the disease, at least temporally, have been reported (9,11).

An autoimmune disease that is often connected with the HBV vaccine is systemic lupus erythematosus (SLE). Numerous cases of SLE manifesting for the first time after immunization against HBV have been described (14-18); some cases have occurred in members of the same family (15). The disease developed in patients of both sexes and all ages (15,18). In many cases, manifestations of the disease included mono- or polyarthritis (14,16).

The possible connection between arthritis (reactive or as part of systemic autoimmune illness) and the recombinant HBV vaccine is important and worthy of further research, because the disease itself is a potentially lethal one, and vaccination against it has become mass practice. We think that special emphasis should be put on possible autoimmune side effects in infants and children, as well as prospective studies in individuals that have been vaccinated against HBV as newborns. Are they more or less prone to develop autoimmune side effects as their system is exposed to these foreign antigens at a very early stage?

Arthritis and Rubella Vaccine

The rubella vaccine has been mentioned often in connection with the onset of reactive arthritis. The wild virus itself has been reported to trigger arthritis in a large number of patients. Similarly, arthritis has also been described in temporal proximity to the administration of the vaccine. During the last 10 years more than 100 cases of rubella vaccine related arthropathies have been reported (19,20). In 1991, the Institute of Medicine released a report in JAMA examining 18 adverse effects of diphtheria-pertussis-tetanus (DPT) vaccine and four adverse effects of rubella vaccine strain 27/3. The report concluded that the evidence suggests a causal relation between rubella vaccine and acute arthritis in adult women (21,22). No animal studies were available to support (or disprove) this conclusion.

In 1997, a study published in JAMA (21) evaluated the risk of persistent joint and neurological symptoms in rubella seronegative women, subsequently vaccinated with rubella vaccine strain RA 27/3. The study was retrospective and included 900 to 1000 women, 15 to 59 years old. No significantly increased risk was associated with receipt of rubella vaccine for any condition, except for carpal tunnel syndrome for vaccinated women at least 30 years old compared to seropositive non-vaccinated women. Six of the seronegative immunized women had onset of symptoms during the 1-year follow-up period. Four of these had transient arthritis or arthralgias (of less than 6 weeks duration), one had arthralgia of indeterminate chronicity, and one had carpal tunnel syndrome (20). The conclusion of this large retrospective study was that no clear-cut evidence indicates increased chronic arthropathies (or neurological symptoms) in women receiving rubella RA 27/3 vaccine. The data supported the continued vaccination of rubella-susceptible women to reduce the risk of congenital rubella syndrome. Postpartum women were less likely to develop non traumatic arthropathies following rubella vaccination.

As with HBV vaccine, a genetic predisposition to develop autoimmune disease may play a role in the appearance of acute reactive arthritis in proximity to the administration of rubella vaccine. In 1998 (23), a group of scientists examined the frequency of HLA class II (HLA-DR) in relation to the incidence of acute arthralgia or arthritis in 283 white women who had received RA 27/3 rubella vaccine (n=146) or placebo (n=137) postpartum. Leukocyte DNA was molecularly typed for HLA-DRB1 gene expression. After statistical analysis of the results, which included adjustment for variables such as age, treatment, time postpartum, the conclusion was that the risk of developing arthropathy was 1.9 times greater after rubella vaccination than after placebo. The risk for arthropathy was also influenced by DR interactions. Odds were 8 times greater in individuals with both DR1 and DR4 and 7.1 times greater with both DR4 and DR6 present, suggesting that coexpression of these specificities may predispose to postpartum arthropathies (23).

Apart from the genetic predisposition discussed above, an additional possible risk factor to post-vaccine arthropathy was examined. Mitchell et al (24) described the results of a study that examined whether pre-vaccination rubella virus specific IgG levels have any correlation with the development of acute or chronic (persistent or recurrent) joint manifestations. Specific IgG levels were determined by whole enzyme immunoassay (EIA), neutralization domain peptide, and neutralization bioassay in prevaccine samples of seronegative women. Of rubella vaccinated women tested for prevaccine antibodies, 21.7% were positive (by EIA) , 17.6% were positive by neutralization domain peptide, and 12.7% had neutralization titers of at least 1:8. Seropositivity tests were significanlty lower in women who developed joint symptoms after immunization compared with those of asymptomatic women. The authors` conclusion was that the risk for arthropathy following RA 27/3 rubella vaccination may be higher in women who have very low prevaccine levels of rubella antibodies, particularly in assays measuring functional (neutralizing) antibodies (24).

Arthritis and Other Vaccines

Fewer reports have appeared on arthritis following other types of vaccines.

Mumps and measles (25), influenza (26), DPT (27), and typhoid (28) are each connected with some cases described during the last 10 years. In the first case, the mumps component of the vaccine was deemed to be responsible for the development of acute monoarthritis (25). In the case of influenza vaccination, the arthritic manifestations were accompanied by neurologic symptoms and signs, namely orbital myosistis and posterior scleritis (26). Indeed, the influenza vaccine has been more often connected to neurologic manifestations (1) than to joint involvement.

Maillefert et al describe two patients who developed monoarthritis after DPT immunization. One patient developed monoarthritis of the knee that resolved only after synovectomy and recurred 5 years later after a booster vaccine injection (27), and the second patient developed monoarthritis of the ankle that persisted for 3 days and resolved spontaneously. These cases suggest that DPT vaccine may cause arthritis, however additional evidence is needed in order to confirm this possibility. An additional vaccine mentioned in connection with arthritis is typhoid. Two cases, both of which were HLA-B27 negative, are reported (28). Reactive arthritis occurred in two travelers who received oral Ty21a typhoid vaccine.

In a previous article (1), we raised the question whether administration of polyvaccines might be more inductive of autoimmune side effects than monovaccines. A series of patients (all of them previously healthy army personnel) have been reported (29) who developed SLE after receiving a number of vaccines among them typhoid, tetanus toxoid, hepatitis B, influenza, and meningococcal vaccine. All patients met the ARC criteria for SLE; all manifested mono- or polyarthritis.

Possible Mechanisms for Post-vaccination Reactive Arthritis

As we mentioned in previous reviews of this subject (1-4), possible mechanisms to explain this phenomenon have been suggested. One of the popular explanations is the "molecular mimicry" mechanism. Certainly in live vaccines, but also in attenuated preparations, antigens of the infective organism remain intact and immunologically potent. Structural similarity between these bacterial or viral antigens and host antigens may enable the triggering of an autoimmune response. In a recent article (30), the author suggests an immunologically based explanation for reactive arthritis (along the lines of molecular mimicry as mentioned above) and possible therapy. He proposes an anti-idiotypic model for the disease. Based on this hypothesis (in support of which evidence is presented in the article) a bacterial lipopolysaccharide (LPS) epitope is recognized by idiotypic (Id) T-cell receptors and antibody Fab immune recognition surfaces (IRS), which have the immunologic appearance of an antigen on the synovial surface. These Id immune effectors utilize an HLA-B27 molecule to present their IRS on their surface, which results in an anti-idiotypic immune response that can also target the synovial antigen. The anti-idiotypic IRS have the immunlogic appearance of LPS and their detection in the arthritic joints falsely suggests the presence of bacterial LPS. The synovial antigen is attacked by the anti-idiotypic response against the bacterial LPS. Therapeutic vaccination is supported by this hypothesis (30).

Additional data supporting the molecular mimcry mechanism are found in a recent study performed on dogs, which, to the best of our knowledge, is the only one of its kind, so far (31). In this study, newborn puppies were immunized with a number of mandatory vaccines (rabies and multivalent vaccine). Based on interpretation of the results, the authors concluded that vaccination did not cause immunosuppression, but it did induce autoantibodies and antibodies to heterologous antigens. None of the dogs developed autoimmune illness, even though the study was discontinued at 22 weeks of age, a time well before any clinical signs of collagen disease usually become apparent. The authors mention similar findings in a follow-up study in dogs that were immunized with the rabies vaccine only and with the multivalent vaccine only.


We have reviewed the existing data on the incidence of acute and chronic arthritis following various immunizations. The vaccines most often connected with the occurrence of reactive arthritis (either isolated or as part of systemic RA) are HBV recombinant vaccine and rubella vaccine. Individuals genetically prone to develop autoimmune illness (namely carriers of HLA-DR1, DR4, DR6, or HLA-B27) are more likely to contract post-immunization arthritis. The serological immune status of rubella seronegative women prior to vaccination also influences the likelihood of acute or chronic arthritis after immunization with rubella vaccine. Immune mechanisms are suggested to explain this phenomenon. A clear causal relation was not proved in any of the cases, however a temporal one apparently exists. There are no means as yet to anticipate which individuals will develop post-vaccination arthritis (other than the genetic tendency detailed above). Vaccination against HBV and rubella, as well as other infectious diseases, no doubt continues - as it should. However, we should be aware of the possibility of autoimmune side effects (in this case specifically joint involvement) to vaccination and perhaps abstain from less than necessary vaccination of patients at known risk for autoimmune illness. These would include people with previously diagnosed disease, with family members suffering from autoimmune disease, or known carriers of genetic risk factors (as detailed above).


Table I. Vaccines associated
with post-vaccination reactive arthritis and RA .



No. of cases

Additional symptoms


Reactive arthritis



Erythema nodosum




Migratory arthritis, urticaria, oedema of the




Hypercalcemia, lytic bone lesions












Adult-onset Still`s disease




Reactive arthritis alone








Mumps and measles



















Rheumatoid arthritis




Reviewd in 1,4

















Reviewed in 1




Cutaneous, renal, hematological



Polyvaccine: mumps, measles, tetanus, meningococcal, hepatitis
A, polio


Cutaneous, renal, hematological



Share this with your friends