Evidence Concerning Rubella Vaccines and Arthritis,
Radiculoneuritis, and Thrombocytopenic Purpura
Clinical Description and Pathologic Aspects
Symptoms referable to the musculoskeletal system are among the most common side effects of rubella and of rubella vaccine. The general term for these symptoms, arthropathy, refers to any abnormality of the joints. Arthropathy encompasses arthralgia (subjective pain in a joint or joints), stiffness (with arthralgia, commonly referred to as rheumatism), and arthritis (objective findings of swelling, redness, heat, or limitation of motion). Arthralgia is more common than arthritis following natural rubella, with both being more common in adults than in children (Lee et al., 1960). Joint symptoms related to natural rubella or rubella vaccine usually begin within 1 week of the appearance of rash in wild-type rubella infection or within 10 to 28 days after immunization. The joints involved, in order of decreasing frequency, are fingers, knees, wrists, elbows, ankles, hips, and toes. The symptoms are frequently of sudden onset and can consist of prominent stiffness and pain only; however, warmth, redness, and effusions occur, especially in the knees, fingers, and wrists (Smith et al., 1987).
The subjective nature of arthralgia makes it a difficult entity to study. Evaluation of reports of arthritis as a possible adverse consequence of rubella vaccine is complicated by the fact that arthritis and arthralgia are commonly subsumed under the heading arthropathy or "joint manifestations."
The prevalence of self-reported arthritis and arthralgia (rheumatism), without regard to cause, in the United States has been estimated from a number of national surveys, including the 1960-1962 National Health Examination Survey (National Center for Health Statistics, 1964), the 1971-1975 National Health and Nutrition Examination Survey (National Center for Health Statistics, 1973, 1978), and the 1987 and 1988 National Health Interview Surveys (National Center for Health Statistics, 1988, 1989). According to the 1988 National Health Interview Survey, approximately 13 percent of respondents surveyed reported currently having "arthritis of any kind or rheumatism." Prevalence rates increased with age, with approximately 0.2 percent of persons under age 18 years and 5.3 percent between ages 18 and 44 years reporting arthritis of any kind or arthralgia. Prevalence rates were higher in women of all ages, with 4.3 percent under age 45 years reporting these conditions in contrast to 2.5 percent of men in the same age group. Rates for whites and blacks under age 45 yearsboth sexes combinedare 3.7 and 2.4 percent, respectively. The combining of arthralgia and arthritis of any kind and the cross-sectional and self-reported nature of National Health Interview Survey data do not permit accurate assessment of the prevalence of chronic or recurrent arthritis in the U.S. population.
History of an Association with Rubella Vaccines
Acute arthralgia and arthritis following vaccination have been noted since the earliest studies of rubella vaccines (Barnes et al., 1972; Cooper et al., 1969; Horstmann et al., 1970; Lerman et al., 1971; Spruance and Smith, 1971; Thompson et al., 1971). These acute events have been associated to various degrees with all rubella vaccine strains and occur more frequently in adult women than in adult men or prepubertal children of either sex (Plotkin, 1988; Polk et al., 1982). Reports of chronic arthropathies following rubella vaccination have been fewer. In 1972, Spruance and colleagues reported recurrent joint symptoms in a group of children receiving one strain of rubella vaccine; however, it was not until the 1980s that more systematic investigation of the possible association of rubella vaccines with chronic arthritis was undertaken (e.g., Cunningham and Fraser, 1985; Tingle et al., 1983, 1985, 1986). The lack of controlled studies, coupled with continued anecdotal reports of chronic arthritis following rubella vaccination (ABC News "20/20" report; J. Hatem, York, Pennsylvania, personal communication, 1990; A. J. Tingle, University of British Columbia, personal communication, 1990), have maintained a level of concern over this possible association.
Evidence from Studies in Humans
Acute Arthropathy and Arthritis
Case Series and Controlled Epidemiologic Studies There is a substantial body of evidence, both from controlled and noncontrolled studies in humans, relating rubella vaccine to acute arthropathy and arthritis. The earliest evidence derives from noncontrolled retrospective and prospective studies conducted in the late 1960s and early 1970s, the former generally designed to test the efficacy of various rubella vaccine strains, rather than their side effects, and the latter as part of the routine administration of the vaccine in population-based immunization campaigns (e.g., Austin et al., 1972; Balfour et al., 1976, 1980; Barnes et al., 1972; Cooper et al., 1969; Dudgeon et al., 1969; Fox et al., 1976; Freestone et al., 1971; Grand et al., 1972; Kilroy et al., 1970; Lerman et al., 1971; Monto et al., 1970; Rowlands and Freestone, 1971; Spruance and Smith, 1971; Swartz et al., 1971; Wallace et al., 1972; Weibel et al., 1972, 1980). The vaccines examined in these studies were the HPV-77 strain (no longer in use), developed in simian tissue culture and then grown for production in dog kidney (DK) culture or in duck embryo (DEV) culture; the Cendehill strain (in limited use now), developed in rabbit kidney tissue culture; and the RA 27/3 strain (most commonly used now), developed in human diploid cells (WI38).
These and more recent studies (e.g., Peltola and Heinonen, 1986; Polk et al., 1982; Valensin et al., 1987) provide generally consistent findings with respect to the acute arthropathy and arthritis observed following rubella immunization. These include the observation that arthropathy and, less commonly, arthritis occur rarely in children but occur in 10 to 40 percent of susceptible (seronegative for rubella at the time of immunization) postpubertal women. Occurrence of these reactions increases with age, and they are less frequent in men and prepubescent children. Rates of acute arthropathy and arthritis following rubella immunization differ by vaccine strain, with the HPV-77 (DK) variant producing the most joint manifestations in all age groups (Barnes et al., 1972; Spruance and Smith, 1971; Wallace et al., 1972). The HPV-77 (DEV) and RA 27/3 strains have also been observed to produce joint symptoms, but the symptoms are more akin to the reaction following natural disease, and the arthropathy is more likely to occur in adults than in children (Polk et al., 1982; Swartz et al., 1971; Weibel et al., 1972).
In one of the few double-masked controlled studies of joint reactions to rubella vaccine conducted to date, Polk and colleagues (1982) compared reactions in 112 adult, seronegative female employees of a Boston hospital receiving either the HPV-77 (DEV) or the RA 27/3 strain in response to a rubella outbreak. Allocation to the vaccines was haphazard, because the
supply of the initial HPV-77 (DEV) vaccine was depleted midway through the immunization program and the RA 27/3 vaccine was substituted in its place. Fifty-nine of the 112 women received HPV-77 (DEV) vaccine and 53 received the RA 27/3 vaccine. Sixty women served as controlsthat is, they received one or the other vaccine, but were seropositive for rubella at the time of vaccination. In the HPV-77 group, 29 percent (17 of 53) reported onset of joint manifestations and 15 percent (9 of 53) reported onset of arthritis within the first 6 weeks following vaccination. In the RA 27/3 group, 26 percent (14 of 53) reported joint manifestations and 11 percent (6 of 53) reported arthritis with onset within 6 weeks after vaccination. Among the controls, 3 percent (2 of 60) reported joint manifestations and none reported arthritis. Joint manifestations in the HPV-77 (DEV) group occurred later and lasted longer than those in the RA 27/3 group, but all cases recovered without sequelae. There were 8 days missed from work in the HPV-77 group, in contrast to 3 days in the RA 27/3 group. Using data from this study and from 22 studies published elsewhere, the authors estimated the frequencies of joint symptoms following administration of the four different vaccine strains to be as shown in Table 7-1.
The relationship of acute joint manifestations to age must be emphasized. Rubella vaccine-associated arthropathies occur rarely in prepubertal children. In one study of the HPV-77 (DEV) vaccine, none of 31 vaccinees under age 13 years experienced joint manifestations, whereas 25 percent (4 of 16) of women in their 20s and 50 percent (9 of 18) of those aged 25 to 33 had such symptoms (Swartz et al., 1971). A later study of the same vaccine strain in a larger number of subjects (Weibel et al., 1972) revealed similar results. None of 276 cases of acute joint manifestations occurred in persons under age 12 years, whereas 4 percent (6 of 157) of persons between ages
TABLE 7-1 Frequencies of Joint Symptoms in Adult Females Following Administration of Four Vaccine Strains
Proportions (%) of Adult Females Developing Acute Arthritis or Arthralgia
95% Confidence Interval
No. of Studies Cited
SOURCE: Adapted from Polk et al. (1982).
12 and 16, 9 percent (12 of 130) of persons between ages 16 and 20, 9 percent (6 of 65) of persons between ages 20 and 25, and 56 percent (14 of 25) of persons older than 25 years did. This study, unlike the previous one by Swartz and colleagues (1971), distinguished between acute arthritis and arthralgia and indicated that the former was rarer, did not occur before age 16 years, and was manifest in only 4 percent (7 of 178) of persons ages 16 through 25 years. Above age 25 years, however, acute arthritis was common, occurring in 46 percent (11 of 24) of women immunized.
In summary, these studies provide consistent evidence that acute arthropathy and, more rarely, acute arthritis can occur following rubella vaccination and that incidence rates are higher in women than in men and increase with age. None of these studies showed an association of rubella vaccines with chronic arthropathies or arthritis, but their generally limited sample sizes and lengths of study follow-up make it unlikely that if there were such an association it could have been detected. Only a few such studies have been conducted to date, and these are described below.
Case Reports and Case Series A case of acute arthritis following natural rubella that progressed to chronic arthritis was reported in 1968 by Martenis and colleagues, and it has been suggested that wild-type rubella virus might play a role in juvenile rheumatoid arthritis (Hart and Marmion, 1977; Martenis et al., 1968; Ogra et al., 1975). These suggestions have raised concern that a possible similar progression of acute to chronic arthritis following rubella immunization might also occur.
Lerman and colleagues (1971) reported a case of persistent arthritis 1 year after vaccination with HPV-77 (DE) strain, and Spruance and colleagues (1972) reported on 11 children suffering recurrent episodes of knee stiffness, sometimes referred to as "catcher's crouch" syndrome, 8 months after receiving HPV-77 (DK) vaccine. This syndrome has been considered to be caused by radiculoneuritis, rather than arthritis, as described later in this chapter. However, follow-up of the children 11, 48, and 66 months after vaccination indicated that 8, 4, and 3 (with 2 lost to follow-up), respectively, of the original 11 children continued to exhibit episodic morning stiffness in the knees (Spruance et al., 1977). One of the cases was evaluated by arthroscopy, and the synovium was found to be hypertrophied posteriorly; culture for rubella virus was negative. Eleven children with recurrent arthritis 36 months after vaccination with HPV-77 (DK) were reported by Thompson and colleagues in 1973, and other cases have been reported since then, some after immunization with the RA 27/3 strain (Tingle et al., 1979a,b, 1984, 1985, 1986; A. Tingle, British Columbia Children's Hospital, personal communication, 1990). Arthritis occurring first within 12 to
21 days after rubella vaccination has been reported to persist for 4 to 7 years after receipt of the HPV-77 (DE) strain, for 2 years in one woman after receipt of the RA 27/3 strain (Tingle et al., 1985), and for 3.5 years in a second woman (Tingle et al., 1984). Two additional young adult women had recurrent arthritis or arthralgia (not otherwise defined) for 18 to 24 months after receipt of the RA 27/3 strain (Tingle et al., 1986).
As part of studies that began about a decade ago into the pathogenesis of the acute arthropathy following rubella immunization and the natural disease, Tingle and colleagues (1983) attempted to correlate arthropathy with specific antibodies to rubella. Seven women with recurrent arthritis were studied retrospectively, and 24 hospital personnel were studied prospectively. The standard assay for assessment of antibodies in the IgG and IgM fractions, hemagglutination inhibition (HAI), and a more sensitive enzymelinked immunosorbent assay (ELISA) were used to study antibody at 6 weeks and 6 months after immunization. There were no differences between those recipients who experienced joint manifestations and those who did not. Later, in 10 of 37 adult women volunteers who were seronegative by HAI testing and who developed acute arthritis after RA 27/3 vaccination, Tingle and colleagues (1983) detected prevaccination antibodies by the ELISA. On the assumption that the standard test (HAI) failed to detect antibodies in some preimmune individuals who tested positive by the more sensitive ELISA, the authors suggested that the acute arthropathy could be the result of a reinfection rather than a primary infection. Pursuing this hypothesis, Tingle and colleagues studied six women with recurrent chronic arthropathy, manifested by polyarticular arthritis beginning 12 days to 3 weeks after the immunization. Chronic arthritis was noted 2 years after vaccination in one woman receiving the RA 27/3 strain and 4 to 7 years after vaccination in three women receiving the HPV-77 (DE) strain. Another woman reported chronic arthralgia more than 6 years following receipt of the HPV-77 (DE) strain. Prevaccination sera of three women in this group were positive by ELISA; the sera of the other women were negative (Tingle et al., 1985).
In a later prospective study, Tingle and colleagues (1986) compared incidence rates in two groups: the first consisted of 23 women and 23 men, ages 11 to 54 years (mean age, 19.2 years), who had natural infection with wildtype rubella virus in 1983, all of whom underwent seroconversion. Arthritis (joint effusion, limitation of movement, heat, or erythema) and arthralgia, both acute and chronic, occurred more frequently following infection with wild-type rubella virus than it did following vaccination. In the women who had the natural infection, 52 percent (12 of 23) had acute arthritis and 13 percent (3 of 23) had arthralgia only. Among the men, 65 percent (15 of 23) had some joint manifestations, 9 percent (2 of 23) had arthritis, and 48 percent (11 of 23) had arthralgia only. These signs and symptoms became manifest within 7 to 10 days after the onset of the rash. Eighteen months
later, 30 percent (7 of 23) of the women had joint manifestations; of these 7 women, 4 had arthritis and 3 had arthralgia. Among the men, 8 percent (2 of 23) had joint manifestations; of these 2 men, 1 each had arthritis or arthralgia. The second group consisted of 44 women students, ages 17 to 33 years (mean age, 23.1 years), given the RA 27/3 vaccine. They were examined at weekly intervals. Within 4 weeks after vaccination, 14 percent (6 of 44) had acute arthritis and 41 percent (18 of 38) had arthralgia only, for a total of 55 percent (24 of 44) of joint manifestations. Eighteen months later, 5 percent (2 of 44) had joint manifestations; it was not specified whether it was arthritis or arthralgia.
In a study of the 1985 epidemic of rubella, Tingle (A. J. Tingle, University of British Columbia, personal communication, 1991) examined 191 seroconverters, of whom 103 were men and 88 were women. Forty-four percent of the women (39 of 88) and 7 percent (7 of 103) of the men had acute arthritis. Twenty-four months later, 30 percent of the women still had joint manifestations; it was not specified whether it was arthritis or arthralgia.
In a retrospective analysis of vaccine reactions to the measles-mumps-rubella (MMR) vaccine administered to 700,000 children in Sweden, Taranger and Wiholm (1984) reported only one case of chronic arthritis. The patient was a 12-year-old boy who developed "juvenile rheumatoid arthritis" 2 years after receiving the vaccine. His HLA haplotype was B27, which has been associated with chronic arthritis. The authors concluded that this case of arthritis was not causally related to the vaccine, but rather was coincidental.
Two hundred eighteen cases of arthritis (ICD 9 code 716.9) occurring within 28 days of immunization with rubella monovalent, measles-rubella (MR), or MMR vaccine were reported through the CDC's MSAEFI system from 1978 to 1990, a period in which approximately 28.8 million doses of rubella vaccines were administered through public mechanisms in the United States (J. Mullen, Centers for Disease Control, personal communication, 1990). Confirmation of these cases in the form of a physician's diagnosis was not required. Of these 218 cases, 43 (20 percent) cases received at least one other vaccine at the time of rubella immunization. One year after immunization, 13 (6 percent) of the original 218 cases were reported as "not recovered"1; an additional 15 cases were lost to follow-up at year 1.
1 Caution is needed in interpreting this category, because cases of arthritis with a comorbidityfor example, otitis mediareported at the time of initial immunization and persisting at 1 year would have been labeled "not recovered," even if the arthritis had resolved. Review of an earlier set of acute arthritis cases subsequently labeled "not recovered" at 1 year indicated that the majority were due to a persisting morbidity other than arthritis (J. Mullen, Centers for Disease Control, personal communication, 1991).
Controlled Epidemiologic Study Polk and colleagues (1982), in the doublemasked, controlled comparison of joint reactions in 112 seronegative women receiving HPV-77 (DEV) or RA 27/3 vaccine described above, reported no cases of chronic or recurrent arthritis or arthralgia. However, the average length of study follow-up was only 6 weeks, and thus, recurrent cases might not have been detected.
Evidence from Studies in Animals
Data from experimental studies in animals are not available, since it has not been possible to develop an animal model for rubella infection.
A suggestion that joint involvement is somehow related to stage of the menstrual cycle has been made, but there is no evidence in support of this hypothesis (Best et al., 1974; Lerman et al., 1971; Swartz et al., 1971).
Pathogenesis of the joint involvement is apparently direct infection of the synovial membrane, inasmuch as the virus has been recovered from the joint fluid of patients with repeated episodes of arthritis for up to 3 to 4 months after vaccination with the HPV-77 strain (Ogra and Herd, 1971). This has also been true in cases of rubella caused by the wild-type strain of the virus (Ogra et al., 1975). Chantler, Tingle, and colleagues have isolated rubella virus from peripheral blood mononuclear leukocytes of several patients with persistent post-rubella vaccine arthritis (Chantler et al., 1981, 1982; Tingle et al., 1985). Two of these patients had received the RA 27/3 strain. In one study, rubella virus was grown from the leukocytes of five of six women with arthritis for up to 5 years after receiving the HPV-77 (DEV) vaccine (Chantler et al., 1982). In a separate study, rubella virus was isolated from mononuclear leukocytes from blood, synovial fluid, or both in 7 of 19 children with chronic arthritis of unknown cause (Chantler et al., 1985). There was no history of recent rubella infection or rubella vaccination, although most of the patients had received HPV-77 (DEV) rubella vaccine in the past. The infecting rubella strain (vaccine versus wild type) was not identified and has not been identified in the other subjects reported to have arthropathy after rubella vaccination.
The role of circulating immune complexes (CICs) in the pathogenesis of rubella arthritis has been suggested in one study, in which 11 of 33 (33 percent) children with postvaccinal arthritis had CICs containing rubella antigen, in contrast to only 3 of the 19 (16 percent) who did not experience arthralgia (Coyle et al., 1982). In a study of adults44 women vaccinated with the RA 27/3 strain and 23 men and 23 women convalescing from wild-
type rubella infectionno statistically significant differences were found in the CICs between those who had joint manifestations and those who did not. The authors concluded that their data ''do not support a direct role for raised CIC levels in the pathogenesis of rubella-associated arthritis or arthralgia" (Singh et al., 1986, p. 115).
Blood lymphocytes from 15 children who had acute arthralgia or arthritis after rubella vaccination exhibited depressed transformation responses to rubella virus compared with the responses of lymphocytes from controls who had no complications following vaccination. This finding suggests that arthritis could result from a selective depression in the subject's cell-mediated immunity at the time of the initial encounter with the virus (Chiba et al., 1976). On the other hand, the lymphocyte responses to rubella antigen in six adult women who had had recurrent arthritis for at least 9 months after HPV-77 (DEV) rubella vaccination were relatively elevated (Ford and Tingle, 1980; Tingle et al., 1983). The antigens used in the two populations may have differed. However, the two reports are not necessarily at variance, because normal or accentuated cell-mediated immunity to the virus would be expected over the course of an intermittent, recurrent infection in an otherwise normal individual. In any event, it is not clear at this time whether patients who develop arthritis, acute or persistent, after rubella vaccination have a specific immune system defect that prevents their systems from clearing the virus normally.
Acute Arthropathy and Arthritis
The body of evidence concerning the possible relation between rubella vaccine and acute joint manifestations includes a number of case series and experimental studies comparing different rubella vaccine strains. These studies indicate a consistent, direct relation of all rubella vaccine strains both to acute arthropathy and to acute arthritis, with the discontinued HPV77 vaccines being associated with the highest reaction rates. Incidence rates following administration of the current RA 27/3 vaccine strain to adult women average 13 to 15 percent. Average rates are much lower in children, increase with age, and are higher in women than in men. Both acute arthropathy and acute arthritis are conditions whose diagnoses are reasonably reliable; thus, the vaccine exposure probably truly preceded onset of these adverse events in most cases. Although acute joint manisfestations can result from a number of different causes, their association with natural rubella infection and the recovery of rubella virus from the joint fluid of persons experiencing acute arthropathy or arthritis support the biologic plausibility of a relation with attenuated rubella virus vaccine.
The body of evidence concerning the possible relation between vaccination with MMR vaccine or its rubella component and chronic arthritis is limited. There are several case reports and case series, one comparative case series, and one double-masked controlled study. Most of the patients described in these publications had received the HPV-77 (DEV) strain of vaccine. Three or four cases of chronic arthritis following immunization with the RA 27/3 strain have been reported in the literature. A large number of unconfirmed cases of chronic arthritis following administration of RA 27/3 strain have been reported to a single institution. All of the cases were adult females.
In the comparative case series, Tingle and colleagues (1986) reported incidence rates of joint manifestations in 67 young adult women following either natural infection with wild-type rubella virus or receipt of the RA 27/ 3 strain. They reported recurrent arthropathy more than 18 months after either natural infection or vaccination, with incidence rates of 30 and 5 percent (two patients), respectively. The latter two patients had "arthritis or arthralgia" that were not otherwise described (Tingle et al., 1986, p. 113). In a double-masked study of 112 women receiving one of two rubella vaccine strains, Polk and colleagues (1982) reported no cases of chronic or recurrent arthritis or arthralgia. However, the length of the follow-up was only 6 weeks, and cases that might have recurred later would not have been detected. In both studies, the nature of the comparison groups precludes estimation of relative risks for exposed versus nonexposed groups.
One of the women with chronic arthritis after vaccination with the RA 27/3 strain had rubella virus isolated from blood mononuclear cells and from breast milk 7 to 9 months postvaccination. A second woman who reported recurrent arthralgia 2 years and 9 months after receipt of the RA 27/3 vaccine also had rubella virus isolated from peripheral blood leukocytes (Tingle et al., 1985).
The current lack of understanding of the natural history and multiple causes of arthritis and the lack of distinction between cases of arthralgia and arthritis in some reports diminish the specificity of the putative association. It is also difficult to establish whether the vaccination truly preceded the adverse event in many cases. The association of chronic arthritis with natural rubella infection and the recovery of a rubella virus from leukocytes from peripheral blood and synovial fluid of women with prolonged arthritis following rubella vaccination suggest, however, a biologically plausible relation between rubella vaccine and chronic arthritis. Moreover, a few cases have been documented in which arthritis, which was ultimately judged to be chronic, began 2 to 3 weeks after vaccination, which is the incubation period of natural rubella and the time period in which the vaccine strain can be isolated (beginning about 1 week after injection).
The evidence indicates a causal relation between the currently used rubella vaccine strain (RA 27/3) and acute arthritis. Incidence rates are estimated to average 13 to 15 percent among adult women following RA 27/3 immunization with much lower levels noted among children, adolescents, and adult men.
The evidence is consistent with a causal relation between the currently used rubella vaccine strain (RA 27/3) and chronic arthritis in adult women, although the evidence is limited in scope and confined to reports from one institution.2 Prospective, double-masked, controlled trials in which subjects are followed for at least 12 months after rubella vaccination are needed to establish this biologically plausible relation. Additional elements of a definitive study would include attempts to isolate rubella virus from the synovial fluid of affected joints of cases of arthropathy and arthritis and molecular-genetic analysis to determine whether the isolated strain is the one that was injected, a strain derived from the injected strain, or an unrelated strain.
RADICULONEURITIS AND OTHER NEUROPATHIES
Radiculoneuritis is a convenient term used to describe a combination of peripheral neuropathy with dorsal root (spinal nerve) pain. The "catcher's crouch" syndrome, which is knee pain that is somewhat relieved by crouching, has been suggested as an example. This and other related conditions, including polyneuropathy, paresthesias, and carpal tunnel syndrome, occur sporadically in association with natural rubella infection (Bailey, 1962; Brodribb, 1963; Courtenay, 1962; Haire and Hadden, 1970; Heathfield, 1962; Hodges, 1940; Moylan-Jones and Penny, 1962; Witney, 1940).
Radiculoneuritis is uncommon, but it can occur at any age. Incidence rates of related neurologic conditions have been published for defined populations (Beghi et al., 1982, 1985; Stevens et al., 1988).
2 When all rubella vaccine strains, including the discontinued HPV-77 (DK) and HPV-77 (DEV) strains, are considered as a group, the evidence indicates a causal relation with chronic arthritis in adult women.
Evidence from Studies in Humans
Case Reports and Case Series
There have been reports of cases of paresthesias (Chin et al., 1971; MortonKute, 1985; Tingle et al., 1985) and pain involving the arms and the knees, the latter giving rise to the "catcher's crouch" syndrome (Deinard et al., 1973; Spruance et al., 1972; Thompson et al., 1971) following rubella vaccination. Three of these occurred after receipt of the RA 27/3 strain (MortonKute, 1985; Tingle et al., 1985). Other related conditions, reported in individual cases, are carpal tunnel syndrome following both HPV-77 (DK) strain (Chin et al., 1971; Thompson et al., 1971) and RA 27/3 strain (Tingle et al., 1985), two cases of optic neuritis following administration of the HPV-77 vaccine in one case (Kazarian and Gager, 1978) and an unstated rubella vaccine in the other (Kline et al., 1982), three cases of transverse myelitis after Cendehill vaccine, a vaccine of unknown rubella strain, and one unstated vaccine (Behan, 1977; Holt et al., 1976), and two cases of Guillain-Barrè syndrome after MMR vaccine that included the HPV-77 (DEV) strain (Gunderman, 1973).
A case of peripheral neuropathy following rubella immunization was reported in 1984 by Taranger and Wiholm (1984). The issue of peripheral neuropathies related to rubella vaccines was reviewed in an article by Schaffner and colleagues (1974), who examined reports of 299 cases. The study concentrated on 32 children with neuropathies who were followed for up to 32 months. Twenty of these children recovered fully and had no recurrences; 10 had minor complaints that persisted during the follow-up period, and 2 developed recurrences of the "catcher's crouch" syndrome after 32 and 33 weeks, respectively. The majority of the 299 total cases and the 32 cases reviewed in detail followed administration of the HPV-77 (DK) vaccine. The authors estimated the rate of these complications to be 2.2 per 1,000 doses and 0.1 per 1,000 doses for the HPV-77 (DK) and the HPV-77 (DEV) strains, respectively.
One case of radiculitis (ICD 9 code 723.4) occurring within 28 days of immunization with MR vaccine was reported through the CDC's MSAEFI system from 1978 to 1990, a period in which approximately 28.8 million doses of rubella-containing vaccine were administered through public mechanisms in the United States (J. Mullen, Centers for Disease Control, personal communication, 1990). No follow-up of the case was made, and a physician's diagnosis was not required.
The body of evidence concerning the possible relation between immunization with monovalent rubella or MMR vaccine and radiculoneuritis and
other neuropathies is limited to case reports and case series. Frequencies of peripheral neuropathies have been estimated from case series to be 2.2 and 0.1 per 1,000 doses of HPV-77 (DK) and HPV-77 (DEV) vaccines, respectively. The lack of comparison groups in these studies precludes estimation of the relative risks of neuropathies in relation to these strains. For the RA 27/3 strain currently in use in the United States, rates of radiculoneuritis and other neuropathies following its receipt are not available. Three cases have been reported after administration of this vaccine strain. Evidence for biologic plausibility consists of the observation that radiculoneuritis and other neuropathies can occur, though uncommonly, after natural infection with wild-type rubella virus.
There is insufficient evidence to indicate a causal relation between the currently used rubella vaccine (RA 27/3) and radiculoneuritis and other neuropathies.
Thrombocytopenic purpura presents as petechiae, purpura, or mucosal bleeding secondary to decreased numbers of platelets in the blood. It has been reported in association with congenital and acquired rubella (Heggie and Robbins, 1969; Morse et al., 1966). The estimated incidence is 1 case of thrombocytopenia in 3,000 cases of natural rubella (Bayer et al., 1965). This complication is not unique to natural rubella, because it is estimated that 70 percent of cases of thrombocytopenia follow various viral illnesses (Cohn, 1976). Possible mechanisms for virus-induced thrombocytopenic purpura include generation of antibodies to a viral antigen that cross-react with some similar antigen on the platelet, resulting in platelet destruction (Baldini, 1966). The virus itself or immune complexes that include the virus might damage or otherwise modify the platelet surface, making it susceptible to removal by the spleen. The contribution of antibody-mediated platelet dysfunction to bleeding in patients with idiopathic thrombocytopenic purpura, however, remains to be established (George and Shattil, 1991).
Thrombocytopenic purpura can occur at any age. No population-based incidence or prevalence rates were identified.
Evidence from Studies in Humans
Evidence for thrombocytopenic purpura following rubella immunization is rarer than that for natural infection and is limited to isolated case reports. Bartos (1972) described a case of thrombocytopenic purpura in a 26-year-old female immunized with a monovalent vaccine. Four other cases of thrombocytopenic purpura have been reported in a 1-year-old girl (Sharma, 1973), a 16-month-old boy and a 16-month-old girl, both of whom received the RA 27/3 strain (Azeemuddin, 1987), and an 18-month-old girl (Neiderud, 1983) following administration of the combined MMR vaccine. All five patients, who were asymptomatic for viral illness prior to immunization, developed petechiae, purpura, or mucosal bleeding 10 days to 3 weeks postimmunization. Platelets were markedly reduced, but peripheral red and white blood cell counts and morphology were normal. In the patients tested, megakaryocytes were usually increased, with an otherwise normal bone marrow. In all cases, the postimmunization thrombocytopenia was transient. No virus-containing immune complexes or anti-platelet antibodies were demonstrated in these cases.
Twenty-six cases of thrombocytopenic purpura occurring within 28 days of immunization with rubella-containing vaccines were reported through the CDC's MSAEFI system from 1978 to 1990, a period in which approximately 28.8 million doses of rubella vaccine were administered through public mechanisms in the United States (J. Mullen, Centers for Disease Control, personal communication, 1990). They were reported as idiopathic (ICD 9 code 287.3). All but 1 of the 26 cases received at least one other antigen at the time of rubella immunization. No follow-up of the cases was made, and a physician's diagnosis was not required.
The report "Compensation for Vaccine-Related Injuries" (Office of Technology Assessment, 1980) lists thrombocytopenia as a "possible" adverse event after immunization, but cites no reference or evidence for this conclusion. The Meruvax II (Rubella Virus Vaccine Live, Merck Sharp & Dohme) package insert states, ''In view of the decreases in platelet counts that have been reported, thrombocytopenic purpura is a theoretical hazard," but it also cites no evidence for this statement (Merck Sharp & Dohme, West Point, Pennsylvania).
The body of evidence concerning the possible relation between vaccination with the RA 27/3 rubella vaccine strain and thrombocytopenic purpura is limited to two or three cases reported in the literature and unconfirmed
cases reported through the CDC's MSAEFI system. The unambiguous clinical presentation of thrombocytopenic purpura would suggest that the vaccine exposure truly preceded the event. The relation is biologically plausible because thrombocytopenic purpura is believed to occur rarely as a complication of natural rubella infection.
There is insufficient evidence to indicate a causal relation between the currently used rubella vaccine (RA 27/3) and thrombocytopenic purpura.
Austin SM, Altman R, Barnes EK, Dougherty WJ. 1972. Joint reactions in children vaccinated against rubella. I. Comparison of two vaccines. American Journal of Epidemiology 95:53-58.
Azeemuddin S. 1987. Thrombocytopenia purpura after combined vaccine against measles, mumps, and rubella. Clinical Pediatrics 26:318.
Bailey G. 1962. Carpal-tunnel syndrome (letter). British Medical Journal 1:1207.
Baldini M. 1966. Idiopathic thrombocytopenic purpura. New England Journal of Medicine 274:1245-1251.
Balfour HH, Balfour CL, Edelman CK, Rierson PA. 1976. Evaluation of Wistar RA27/3 rubella virus vaccine in children. American Journal of Diseases of Children 130:1089-1091.
Balfour HH, Groth KE, Edelman CK. 1980. RA 27/3 rubella vaccine. American Journal of Diseases of Children 134:350-353.
Barnes EK, Altman R, Austin SM, Dougherty WJ. 1972. Joint reactions in children vaccinated against rubella. Study II: Comparison of the three vaccines. American Journal of Epidemiology 95:59-66.
Bartos HR. 1972. Thrombocytopenia associated with rubella vaccination. New York State Journal of Medicine 72:499.
Bayer WL, Sherman FE, Michaels RH, Szeto ILF, Lewis JH. 1965. Purpura in congenital and acquired rubella. New England Journal of Medicine 273:1362-1366.
Beghi E, Kurland LT, Mulder DW. 1982. Incidence of acute transverse myelitis in Rochester, Minnesota, 1970-1980, and implications with respect to influenza vaccine. Neuroepidemiology 1:176-188.
Beghi E, Kurland LT, Mulder DW, Nicolosi A. 1985. Brachial plexus neuropathy in the population of Rochester, Minnesota, 1970-1981. Annals of Neurology 18:320-323.
Behan PO. 1977. Diffuse myelitis associated with rubella vaccination (letter). British Medical Journal 1:166.
Best JM, Banatvala JE, Bowen JM. 1974. New Japanese rubella vaccine: comparative trials. British Medical Journal 3:221-224.
Brodribb HS. 1963. Rubella (1962) in three boarding schools: its complications and types. British Journal of Clinical Practice 17:726-730.
Chantler JK, Ford DK, Tingle AJ. 1981. Rubella-associated arthritis: rescue of rubella virus from peripheral blood lymphocytes two years postvaccination. Infection and Immunity 32:1274-1280.
Chantler JK, Ford DK, Tingle A. 1982. Persistent rubella infection and rubella-associated arthritis. Lancet 1:1323-1325.
Chantler JK, Tingle AJ, Petty RE. 1985. Persistent rubella infection with chronic arthritis in children. New England Journal of Medicine 313:1117-1123.
Chiba Y, Sadeghi E, Ogra P. 1976. Abnormalities of cellular immune response in arthritis induced by rubella vaccination. Journal of Immunology 177(5, part 1):1684-1687.
Chin J, Werner SB, Kusumoto HH, Lennette EH. 1971. Complications of rubella immunization in children. Western Journal of Medicine 114:7-12.
Cohn J. 1976. Thrombocytopenia in childhood: an evaluation of 433 patients. Scandinavian Journal of Hematology 16:226-240.
Cooper LZ, Ziring PR, Weiss HJ, Matters BA, Krugman S. 1969. Transient arthritis after rubella vaccination. American Journal of Diseases of Children 118:218-225.
Courtenay MJF. 1962. Carpal-tunnel syndrome (letter). British Medical Journal 1:1698.
Coyle PK, Wolinsky JS, Buimovici-Klein E, Moucha R, Cooper LZ. 1982. Rubella-specific immune complexes after congenital infection and vaccination. Infection and Immunity 36:498-503.
Cunningham AL, Fraser JRE. 1985. Persistent rubella virus infection of human synovial cells cultured in vitro. Journal of Infectious Diseases 151:638-645.
Deinard AS, Hoban TW, Venters HD. 1973. Clinical reactions in children after rubella vaccination. Health Services Reports 88:457-462.
Dudgeon JA, Marshall WC, Peckham CS. 1969. Rubella vaccine trials in adults and children. American Journal of Diseases of Children 118:237-242.
Ford DK, Tingle AJ. 1980. Cell-mediated immune responses in patients with recurrent arthritis following rubella immunization. Journal of Rheumatology 7:225-230.
Fox JP, Rainey HS, Hall CE, Ray CG, Patterson MJ. 1976. Rubella vaccine in postpubertal women: experience in western Washington state. Journal of the American Medical Association 236:837-843.
Freestone DS, Prydie J, Hamilton-Smith SG, Laurence G. 1971. Vaccination of adults with Wistar RA 27/3 rubella vaccine. Journal of Hygiene 69:471-477.
George JN, Shattil SJ. 1991. The clinical importance of acquired abnormalities of platelet function. New England Journal of Medicine 324:27-39.
Grand MG, Wyll SA, Gehlbach SH, Landrigan PJ, Judelsohn RG, Zendel SA, Witte JJ. 1972. Clinical reactions following rubella vaccination. Journal of the American Medical Association 220:1569-1572.
Gunderman JR. 1973. Guillain-Barrè syndrome: occurrence following combined mumps-rubella vaccine. American Journal of Diseases of Children 125:834-835.
Haire M, Hadden DSM. 1970. Immunoglobulin responses in rubella and its complications. British Medical Journal 3:130-132.
Hart H, Marmion BP. 1977. Rubella virus and rheumatoid arthritis. Arthritis and Rheumatism 5:378-383.
Heathfield KWG. 1962. Carpal-tunnel syndrome (letter). British Medical Journal 2:58.
Heggie AD, Robbins FC. 1969. Natural rubella acquired after birth: clinical features and complications. American Journal of Diseases of Children 118:12-17.
Hodges GMW. 1940. Brachial neuritis following rubella (letter). British Medical Journal 1:548.
Holt S, Hudgins D, Krishnan KR, Critchley EMR. 1976. Diffuse myelitis associated with rubella vaccination. British Medical Journal 2:1037-1038.
Horstmann DM, Liebhaber H, Kohorn EI. 1970. Post-partum vaccination of rubella-susceptible women. Lancet 2:1003-1006.
Kazarian EL, Gager WE. 1978. Optic neuritis complicating measles, mumps and rubella vaccination. American Journal of Ophthalmology 86:544-547.
Kilroy AW, Schaffner W, Fleet WF, Lefkowitz LB, Karzon DT, Fenichel GM. 1970. Two syndromes following rubella immunization: clinical observations and epidemiological studies. Journal of the American Medical Association 214:2287-2292.
Kline LB, Margulies SL, Oh SJ. 1982. Optic neuritis and myelitis following rubella vaccination. Archives of Neurology 39:443-444.
Lee PR, Barnett AF, Scholer JF, Bryner S, Clark WH. 1960. Rubella arthritis: a study of twenty cases. California Medicine 93:125-128.
Lerman SJ, Nankervis GA, Heggie AD, Gold E. 1971. Immunologic response, virus excretion, and joint reactions with rubella vaccine: a study of adolescent girls and young women given live attenuated virus vaccine (HPV-77:DE-5). Annals of Internal Medicine 74:67-73.
Martenis TW, Bland JH, Phillips CA. 1968. Rheumatoid arthritis after rubella. Arthritis and Rheumatism 11:683-687.
Monto AS, Cavallaro JJ, Whale EH. 1970. Frequency of arthralgia in women receiving one of three rubella vaccines. Archives of Internal Medicine 126:635-639.
Morse EE, Zinkham WH, Jackson DP. 1966. Thrombocytopenic purpura following rubella infection in children and adults. Archives of Internal Medicine 117:573-579.
Morton-Kute L. 1985. Rubella vaccine and facial parasthesias (letter). Annals of Internal Medicine 102:563.
Moylan-Jones RJ, Penny PT. 1962. Complications of rubella (letter). Lancet 2:355.
National Center for Health Statistics. 1964. Cycle I of the National Health Examination Survey, Sample and Response, United States, 1960-62. Vital and Health Statistics, Series 11, No. 1. PHS Publication No. 1000. Public Health Service. Washington, DC: U.S. Government Printing Office.
National Center for Health Statistics. 1973. Plan and Operation of the National Health and Nutrition Examination Survey, United States, 1971-73. Vital and Health Statistics, Series 1, Nos. 10a and 10b. DHEW Publication No. (HSM) 73-1310. Health Services and Mental Health Administration. Washington, DC: U.S. Government Printing Office.
National Center for Health Statistics. 1978. Plan and Operation of the NHANES I Augmentation Survey of Adults 25-74 Years, United States, 1974-75. Vital and Health Statistics, Series 1, No. 14. DHEW Publication No. (PHS) 78-1314. Public Health Service. Washington, DC: U.S. Government Printing Office.
National Center for Health Statistics. 1988. Current Estimates from the National Health Interview Survey, United States, 1987. Vital and Health Statistics, Series 10, No. 166. DHHS Publication No. (PHS) 88-1594. Public Health Service. Washington, DC: U.S. Government Printing Office.
National Center for Health Statistics. 1989. Current Estimates from the National Health Interview Survey, United States, 1988. Vital and Health Statistics, Series 10, No. 173. DHHS Publication No. (PHS) 89-1501. Public Health Service. Washington, DC: U.S. Government Printing Office.
Neiderud J. 1983. Thrombocytopenic purpura after a combined vaccine against morbilli, parotitis and rubella. Acta Paediatrica Scandinavica 72:613-614.
Office of Technology Assessment. 1980. Compensation for vaccine-related injuries: a technical memorandum. Washington, DC: U.S. Government Printing Office.
Ogra PL, Herd JK. 1971. Arthritis associated with induced rubella infection. Journal of Immunology 107:810-813.
Ogra PL, Chiba Y, Ogra SS, Dzierba JL, Herd JK. 1975. Rubella-virus infection in juvenile rheumatoid arthritis. Lancet 1:1157-1161.
Peltola H, Heinonen OP. 1986. Frequency of true adverse reactions to measles-mumps-rubella vaccine: a double-blind placebo-controlled trial in twins. Lancet 1:939-942.
Plotkin SA. 1988. Rubella vaccine. In: Plotkin SA, Mortimer EA, eds. Vaccines. Philadelphia: W.B. Saunders Co.
Polk BF, Modlin JF, White JA, DeGirolami PC. 1982. A controlled comparison of joint reactions among women receiving one of two rubella vaccines. American Journal of Epidemiology 115:19-25.
Rowlands DF, Freestone DS. 1971. Vaccination against rubella of susceptible girls in Reading. Journal of Hygiene 69:579-586.
Schaffner W, Fleet WF, Kilroy AW, Lefkowitz LB, Herrmann KL, Thompson J, Karzon DT. 1974. Polyneuropathy following rubella immunization: a follow-up study and review of the problem. American Journal of Diseases of Children 127:684-688.
Sharma ON. 1973. Thrombocytopenia following measles-mumps-rubella vaccination in a one-year-old infant. Clinical Pediatrics 12:315.
Singh VK, Tingle AJ, Schulzer M. 1986. Rubella-associated arthritis. II. Relationship between circulating immune complex levels and joint manifestations. Annals of the Rheumatic Diseases 45:115-119.
Smith CA, Petty RE, Tingle AJ. 1987. Rubella virus and arthritis. Rheumatic Disease Clinics of North America 13:265-274.
Spruance SL, Smith CB. 1971. Joint complications associated with derivatives of HPV-77 rubella virus vaccine. American Journal of Diseases of Children 122:105-111.
Spruance SL, Klock LE, Bailey A, Ward JR, Smith CB. 1972. Recurrent joint symptoms in children vaccinated with HPV-77DK12 rubella vaccine. Journal of Pediatrics 80:413-417.
Spruance SL, Metcalf R, Smith CB, Griffiths MM, Ward JR. 1977. Chronic arthropathy associated with rubella vaccination. Arthritis and Rheumatism 20:741-747.
Stevens JC, Sun S, Beard CM, O'Fallon WM, Kurland LT. 1988. Carpal tunnel syndrome in Rochester, Minnesota, 1961 to 1980. Neurology 38:134-138.
Swartz TA, Klingberg W, Goldwasser RA, Klingberg MA, Goldblum N, Hilleman MR. 1971. Clinical manifestation according to age, among females given HPV-77 duck rubella vaccine. American Journal of Epidemiology 94:246-251.
Taranger J, Wiholm BE. 1984. Litet antal biverkningar rapporterade efter vaccination mot massling-passjuka-roda hund. Lakartidningen 81:1936-1937.
Thompson GR, Ferreyra A, Brackett RG. 1971. Acute arthritis complicating rubella vaccination. Arthritis and Rheumatism 14:19-26.
Thompson GR, Weiss JJ, Shillis JL, Brackett RG. 1973. Intermittent arthritis following rubella vaccination: a three year follow-up. American Journal of Diseases of Children 125:526-530.
Tingle AJ, Ford DK, Price GE, Kettyls DWG. 1979a. Prolonged arthritis in identical twins after rubella immunization (brief report). Annals of Internal Medicine 90:203-204.
Tingle AJ, Kettyls GDM, Ford DK. 1979b. Studies on vaccine-induced rubella arthritis: serologic findings before and after immunization. Arthritis and Rheumatism 22:400-402.
Tingle AJ, Yang T, Allen M, Kettyls GD, Larke RP, Schulzer M. 1983. Prospective immunological assessment of arthritis induced by rubella vaccine. Infection and Immunity 40:22-28.
Tingle AJ, Pot CH, Chantler JK. 1984. Prolonged arthritis, viraemia, hypogammaglobulinaemia, and failed seroconversion following rubella immunisation (letter). Lancet 1:1475-1476.
Tingle AJ, Chantler JK, Pot KH, Paty DW, Ford DK. 1985. Postpartum rubella immunization: association with development of prolonged arthritis, neurological sequelae, and chronic rubella viremia. Journal of Infectious Diseases 152:606-612.
Tingle AJ, Allen M, Petty RE, Kettyls GD, Chantler JK. 1986. Rubella-associated arthritis. I. Comparative study of joint manifestations associated with natural rubella infection and RA27/3 rubella immunisation. Annals of the Rheumatic Diseases 45:110-114.
Valensin PE, Rossolini GM, Cusi MG, Zanchi A, Cellesi C, Rossolini A. 1987. Specific antibody patterns over a two-year period after rubella immunization with RA 27/3 live attenuated vaccine. Vaccine 5:289-294.
Wallace RB, Libert P, Ibrahim M, Isacson P. 1972. Joint symptoms following an area-wide rubella immunization campaign: report of a survey. American Journal of Public Health 62:658-661.
Weibel RE, Stokes J, Buynak EB, Hilleman MR. 1972. Influence of age on clinical response to HPV-77 duck rubella vaccine. Journal of the American Medical Association 222:805-807.
Weibel RE, Villarejos VM, Klein EB, Buynak EB, McLean AA, Hilleman MR. 1980. Clinical and laboratory studies of live attenuated RA 27/3 and HPV 77-DE rubella virus vaccines. Proceedings of the Society for Experimental Biology and Medicine 165:44-49.
Witney EW. 1940. Neuritis following rubella (letter). British Medical Journal 1:831.