
* Hospital for Children and Adolescents, Helsinki University Central Hospital, Helsinki, Finland
Department of Infectious Disease Epidemiology, National Public Health Institute, Helsinki, Finland
| ABSTRACT |
|---|
|
|
|---|
Methods. A retrospective study based on linkage of individual MMR vaccination data with a hospital discharge register was conducted among 535 544 1- to 7-year-old children who were vaccinated between November 1982 and June 1986 in Finland. For encephalitis and aseptic meningitis, the numbers of events observed within a 3-month risk interval after vaccination were compared with the expected numbers estimated on the basis of occurrence of encephalitis and aseptic meningitis during the subsequent 3-month intervals. Changes in the overall number of hospitalizations for autism after vaccination throughout the study period were searched for. In addition, hospitalizations because of inflammatory bowel diseases were checked for the children with autism.
Results. Of the 535 544 children who were vaccinated, 199 were hospitalized for encephalitis, 161 for aseptic meningitis, and 352 for autistic disorders. In 9 children with encephalitis and 10 with meningitis, the disease developed within 3 months of vaccination, revealing no increased occurrence within this designated risk period. We detected no clustering of hospitalizations for autism after vaccination. None of the autistic children made hospital visits for inflammatory bowel diseases.
Conclusions. We did not identify any association between MMR vaccination and encephalitis, aseptic meningitis, or autism.
Key Words: measles mumps rubella MMR vaccine immunization adverse ef-fects encephalitis aseptic meningitis autism autistic dis-orders
Abbreviations: MMR, measles-mumps-rubella MIBE, measles inclusion body encephalitis SSPE, subacute sclerosing panencephalitis CSF, cerebrospinal fluid ICD, International Classification of Diseases
| INTRODUCTION |
|---|
|
|
|---|
Measles, mumps, and rubella viruses are neurotropic.1 Involvement of the central nervous system is common in measles, and electroencephalographic changes have been reported in 50% of uncomplicated cases.4 Measles virus causes a variety of central nervous system syndromes, including meningitis,5 encephalitis,5,6 measles inclusion body encephalitis (MIBE),5,6 subacute sclerosing panencephalitis (SSPE),5,6 and acute disseminated encephalomyelitis.1 Acute encephalitis develops in 35 to 100 of 100 000 measles patients. The mortality rate is 10% to 20%, and neurologic damage occurs in 25% of survivors.68
Before the introduction of vaccination, mumps was the most common cause of viral encephalitis in children in several countries.9 The reported incidence of mumps encephalitis averages 260 per 100 000 cases.4,7 Estimates of the rate of clinical meningitis range from 0.1% to 15%, but 50% of mumps patients show pleocytosis of the cerebrospinal fluid (CSF).4,7 With rubella, encephalitis develops in 13 of 100 000 patients.7,10
Electroencephalographic changes without neurologic symptoms have also been reported in children receiving live measles vaccine.11 Cases of meningitis, encephalitis, MIBE, and acute disseminated encephalomyelitis have been reported after MMR vaccinations, but in most cases the link has remained unclear.1,9,1215 An association was suggested on the basis of clustering of cases of encephalitis after vaccination, but the reported rates were indistinguishable from the background rates.8,16 However, MMR vaccines containing the Urabe or the Leningrad-3 strain of mumps virus have been shown to cause meningitis.1720 As a result, Urabe-containing MMR vaccines have been withdrawn from most countries.21
More recently, MMR vaccine has been suggested as 1 reason for the increasing incidence of autistic disorders.22,23
By linking data from hospital discharge and vaccination registers, we assessed whether an association prevails between MMR vaccination and encephalitis, aseptic meningitis, or autism.
| METHODS |
|---|
|
|
|---|
86% of all children scheduled to be vaccinated between November 1982 and June 1986 in Finland.24 M-M-RII (Merck & Co, West Point, PA) was the only vaccine in use in Finland during the enrollment. This vaccine contains the more attenuated Enders-Edmonston strain of measles virus, the Jeryl Lynn strain of mumps virus, and the Wistar RA 27/3 strain of rubella virus.
Hospital Discharge Register
The nationwide hospital discharge register includes data on all hospitalizations since 1972 and has a validated high coverage (over 95%).25 Individual hospitalizations are identified from the register by using social security numbers and the International Classification of Diseases (ICD) codes of the World Health Organization. ICD-8 (effective from 1969 through 1986) and ICD-9 (effective from 1987 through 1995) codes listed in Table 1 were used for case collection.
|
For calculation of the background incidences, hospitalizations among the 1- to 7-year-old children who were not vaccinated during the enrollment were also searched for. For autism, only the first hospital visit during the study period was included in the survey. If acute encephalitis or meningitis caused several hospitalizations in the same child, all visits were assessed. In addition, hospitalizations because of inflammatory bowel diseases during 19821995 were evaluated for the children with autism.
Of the patients hospitalized because of encephalitis or meningitis within 3 months of MMR vaccination, the exact dates of immunization were collected from the patients medical records or personal vaccination cards filed at health centers. For verified other patients, the dates, based on the year and month of vaccination, could be estimated with an accuracy of 1 month. To assess the accuracy of the ICD coding and to evaluate the role of other causes for the events, we reviewed the medical records of all patients hospitalized for encephalitis or meningitis within 3 months of vaccination. Cases meeting the diagnostic criteria listed in Table 2 were further analyzed.
|
Statistical Methods
For encephalitis and aseptic meningitis, we compared the numbers of events observed within the 3-month risk intervals postvaccination with the numbers expected. The numbers expected were calculated on the basis of the numbers of events observed during the subsequent 3-month intervals until 24 months after vaccination. The data were analyzed using the
2 test with the Yates correction.29 P values of < .05 were considered significant.
Because no risk period could be defined for autistic disorders, we evaluated whether there were changes in the overall number of hospitalizations for autism after MMR vaccination during the whole study period.
| RESULTS |
|---|
|
|
|---|
Encephalitis
Of the 199 children, 9 were hospitalized for encephalitis within 3 months of vaccination (Table 3). MMR vaccine was administered to 80 children after the disease, and in 110 the interval between vaccination and hospitalization exceeded 3 months. In addition, 66 events were observed among unvaccinated 1- to 7-year-old children.
|
The incidence of encephalitis of undefined cause among all 1- to 7-year-old children decreased by 35% from 19.90 per 100 000 in 1983 to 13.00 per 100 000 in 1985. The annual numbers of hospitalizations for encephalitis among children in the vaccination register are illustrated in Fig 1 and hospitalizations of unvaccinated 1- to 7-year-old children in Fig 2.
|
|
The incidence of meningitis of undefined cause in 1- to 7-year-old children decreased by 24% during the study period from 10.17 per 100 000 in 1983 to 7.71 per 100 000 in 1985 (absolute numbers in Figs 1 and 2).
Autistic Disorders
Of the vaccinees, 309 were hospitalized for autism after vaccination. When the shortest possible intervals between MMR vaccination and the day of hospitalization were assessed, these ranged from 3 days to 12 years and 5 months. No distinguishable clustering was detected in the intervals from vaccination to the hospitalization. The number of hospital admissions remained relatively steady during the first 3 years and then gradually decreased, as was expected because of the increasing age of the vaccinees (Fig 3). Forty-three children were vaccinated after the first hospitalization and 31 were hospitalized but remained unvaccinated between November 1982 and June 1986.
|
| DISCUSSION |
|---|
|
|
|---|
Several other studies have evaluated the relation between live virus vaccinations and neurologic disorders. During 19631971 in the United States, a clustering of 45 cases of encephalitis was detected 6 to 15 days after measles vaccination. A definite link with the vaccine was not established in any of the cases, but was regarded possible. The incidence of neurologic disorders in the recipients of further attenuated vaccines was estimated as 0.08 per 100 000 doses.16 A Canadian study found a rate of 0.18 cases of encephalitis per 100 000 doses of measles vaccine, which was very close to the background level of encephalitis of unspecified cause.30 Weibel et al8 reported a clustering of 17 cases of encephalopathy on days 8 and 9 after measles, measles-rubella, or MMR vaccination, but the authors stated that, with a denominator of 75 000 000 vaccinees throughout 23 years, encephalopathy would be an extremely rare complication (0.06 per 100 000 vaccinees).
In addition, several case reports of encephalitis occurring after monovalent or combination MMR vaccinations exist, but in most cases causality has not been proved.1,9,12,13 However, the Urabe mumps vaccine strain has been shown to cause encephalitis,31 and measles virus with a nucleotide sequence identical to the more attenuated Enders-Edmonston vaccine strain was isolated from the brain tissue of an immunodeficient patient developing fatal MIBE 8 months after MMR vaccination.15 Development of SSPE has been described 3 weeks after live measles vaccination in a child with no history of measles.32 This probably indicates mere concurrence, because only wild-type measles virus has been isolated from patients with SSPE.1 Reassuringly, measles immunization has dramatically diminished the incidence of SSPE.1
The problem of vaccine-associated meningitis has been prominent with the Japanese MMR vaccines. In several cases, mumps virus has been isolated from CSF and identified by nucleotide sequencing analysis to be the Urabe vaccine strain.31,33 The incidence of meningitis attributable to the Urabe vaccine varies from 3.5 to 166 per 100 000 doses.19,34 A mass immunization campaign with the Urabe-containing MMR vaccine in Brazil resulted in 58 cases of aseptic meningitis. The relative risk 3 weeks postvaccination as compared with the risk before the campaign was 14.3 (95% confidence interval: 7.925.7, 7.1 per 100 000 doses).20 The Leningrad-3 strain of mumps has also been shown by virus isolation to cause meningitis in 90 to 100 per 100 000 vaccine recipients.17
Rare cases of meningitis have been reported after vaccination with the Jeryl Lynn mumps strain, but causality has not been proved in any of the cases.1,13,14 Mumps virus was isolated from the CSF of a child with meningitis occurring 21 days after Jeryl Lynn vaccination, but the virus was not reliably identified as wild or vaccine virus.1,13
In 1998, Wakefield et al22 suggested that MMR vaccine could cause enterocolitis leading to excessive absorption of peptides, disturbance of neurologic development, and autistic disorder within 14 days of immunization. This theory has been rebutted because of several methodological weaknesses, and the contradictory results of subsequent reports.3537 Taylor et al35 investigated by the case series method whether clustering of autism occurred after MMR vaccination and found no support for the hypothesized link. A similar conclusion was reached in 2 time trend analyses from the United Kingdom and the United States.36,37 The incidence of autism varies widely among studies, and the observed increase may reflect better case ascertainment and the use of different definitions for the disorders.38 Although the first symptoms of autism are typically manifested at the age of MMR vaccination, there is no epidemiologic evidence that immunization causes autism.3537
Reliable assessment of causality between immunization and rare disorders is extremely difficult. Therefore, the evidence of several of the suspected adverse effects of MMR vaccination has remained controversial or inconclusive.1 Linkage of MMR vaccination and hospital discharge registers provided us with an opportunity to evaluate these complex issues further, but certain limitations were unavoidable. We had no access to data of outpatient visits. However, the occurrence of severe encephalitis and meningitis requiring hospitalization could be assessed reliably. For children with encephalitis and meningitis, the interval between vaccination and the day of hospitalization was calculated because the exact date of occurrence of symptoms was not always clear. Because these acute diseases usually lead to hospitalization within a few days of the onset of symptoms, excess of illness after immunization would have been detected.
The exact incidence of autism could not be defined with our approach, because autistic disorders develop insidiously over long periods of time, or the disorder is present at birth but not obvious until later, and the first hospitalization does not indicate the timing of the occurrence of symptoms. Furthermore, diagnosis of autism does not always involve hospitalization. However, in Finland it is common that these children are admitted to hospital for observation, in-depth neurobiological examinations, treatment, and rehabilitation. Thus, a significant clustering of hospital admissions for autistic disorders after MMR vaccination would have been detected in this study.
Furthermore, as the coverage of the MMR vaccination register was not complete, some children regarded as unvaccinated may actually have been immunized during the study period. Because the number of unvaccinated children is minuscule as compared with the number of those vaccinated, this limitation is unlikely to influence the findings of this study.
Whether the cases of encephalitis and meningitis occurring within 2 days of vaccination should have been excluded from the analysis, because no viremia is to be expected within such a short interval, is debatable.4 Correspondingly, the designated risk interval of 3 months exceeds the incubation periods of natural measles, mumps, and rubella,4 but was chosen because of the suggestions that attenuation of viruses may prolong the usual incubation periods.12
Our results provide additional evidence of the safety of MMR vaccination. Nevertheless, significant public concern about adverse events of vaccines clearly exists, and continuous surveillance aiming at distinguishing true adverse events from unrelated, chance occurrences is crucial to maintain public confidence in immunization.
| ACKNOWLEDGMENTS |
|---|
We thank the MMR follow-up group for valuable advice and Timo Kanninen and Teemu Möttönen for help with data analysis.
| FOOTNOTES |
|---|
Reprint requests to (A.M.) Hospital for Children and Adolescents, Helsinki University Central Hospital, Stenbäckinkatu 11, PL 281, 00029 HUS, Finland. E-mail: annamari.makela{at}hus.fi
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
K. N. Ward, N. J. Bryant, N. J. Andrews, J. S. Bowley, A. Ohrling, C. M. Verity, E. M. Ross, and E. Miller Risk of Serious Neurologic Disease After Immunization of Young Children in Britain and Ireland Pediatrics, August 1, 2007; 120(2): 314 - 321. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. J. Woo, R. Ball, R. Landa, A. W. Zimmerman, M. M. Braun, and VAERS Working Group, Center for Biologics Evaluati Developmental regression and autism reported to the Vaccine Adverse Event Reporting System Autism, July 1, 2007; 11(4): 301 - 310. [Abstract] [PDF] |
||||
![]() |
Y. D'Souza, E. Fombonne, and B. J. Ward No Evidence of Persisting Measles Virus in Peripheral Blood Mononuclear Cells From Children With Autism Spectrum Disorder Pediatrics, October 1, 2006; 118(4): 1664 - 1675. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Vestergaard, A. Hviid, K. M. Madsen, J. Wohlfahrt, P. Thorsen, D. Schendel, M. Melbye, and J. Olsen MMR Vaccination and Febrile Seizures: Evaluation of Susceptible Subgroups and Long-term Prognosis JAMA, July 21, 2004; 292(3): 351 - 357. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. C Klein and E. B Diehl Relationship Between MMR Vaccine and Autism Ann. Pharmacother., July 1, 2004; 38(7): 1297 - 1300. [Abstract] [Full Text] [PDF] |
||||
![]() |
E. J. Woo, R. Ball, A. Bostrom, S. V. Shadomy, L. K. Ball, G. Evans, and M. Braun Vaccine Risk Perception Among Reporters of Autism After Vaccination: Vaccine Adverse Event Reporting System 1990-2001 Am J Public Health, June 1, 2004; 94(6): 990 - 995. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. G. Gurney, M. S. Fritz, K. K. Ness, P. Sievers, C. J. Newschaffer, and E. G. Shapiro Analysis of Prevalence Trends of Autism Spectrum Disorder in Minnesota Arch Pediatr Adolesc Med, July 1, 2003; 157(7): 622 - 627. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Wilson, E. Mills, C. Ross, J. McGowan, and A. Jadad Association of Autistic Spectrum Disorder and the Measles, Mumps, and Rubella Vaccine: A Systematic Review of Current Epidemiological Evidence Arch Pediatr Adolesc Med, July 1, 2003; 157(7): 628 - 634. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. E. Mullins Measles-Mumps-Rubella Vaccine and Autism Pediatrics, July 1, 2003; 112(1): 206 - 206. [Full Text] [PDF] |
||||
![]() |
MMR vaccine - how effective and how safe? DTB, April 1, 2003; 41(4): 25 - 29. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. H. Rathore and L. L. Barton MMR Vaccine and Autism: A Population-based Study AAP Grand Rounds, February 1, 2003; 9(2): 16 - 17. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||