J Neurol (2011) 258:1197–1206

DOI 10.1007/s00415-011-5984-2

REVIEW

Immunizations and risk of multiple sclerosis: systematic review

and meta-analysis
Mauricio F. Farez • Jorge Correale

Received: 6 January 2011 / Revised: 18 February 2011 / Accepted: 25 February 2011 / Published online: 24 March 2011
Ó Springer-Verlag 2011

Abstract The role of vaccinations in risk of developing MMR, Polio and Typhoid fever immunization, whereas
multiple sclerosis (MS) or in risk of relapse has not been diphtheria and tetanus vaccination may be associated with a
well established. The aim of this study was to estimate the decreased risk of MS. Further research is needed for the
effect of immunizations on risk of developing MS in adults remaining vaccines.
as well as in subsequent risk of relapse. Systematic search
for publications in MEDLINE (1966–January 2011), EM- Keywords Multiple sclerosis
Systematic review

BASE (1977–January 2011) and the Cochrane Central Meta-analysis
Vaccines
Register of Controlled Trials (CENTRAL) (1961–January
2011). Both randomized clinical trials and non-randomized
studies addressing the effect of any Center for Diseases Introduction
Control (CDC) recommended vaccine for children, adults
or travelers and BCG on risk of MS or disease relapse were Widespread use of immunizations has been one of the most
included. Two reviewers independently extracted informa- successful measures adopted in public health [1]. Despite
tion from articles selected using a predefined datasheet. No the overwhelming benefits reported for most vaccines,
significant change in the risk of developing MS after vac- concerns are raised from time to time linking immuniza-
cination was found for BCG (OR 0.96, 95% CI 0.69–1.34), tions to different conditions such as autism [2] and auto-
Hepatitis B (OR 1.00, 95% CI 0.74–1.37), Influenza (OR immune disorders [3] such as multiple sclerosis (MS) [4].
0.97, 95% CI 0.77–1.23), Measles–Mumps–Rubella Unfortunately, in many cases, these studies are merely case
(MMR) (OR 1.02, 95% CI 0.64–1.61), Polio (OR 0.87, 95% reports, poorly designed observational studies, or well-
CI 0.61–1.25) and Typhoid fever (OR 1.05, 95% CI designed studies with a small number of participants from
0.72–1.53). We found decreased risk of developing MS for which valid conclusions cannot be drawn. Thus, estab-
Diphtheria (OR 0.60, 95% CI 0.40–0.90) and Tetanus (OR lishing positive, negative or lack of association between the
0.68, 95% CI 0.54–0.84). Influenza immunization was also risk of developing MS or an exacerbation of the disease
associated with no change in risk of MS relapse (RR 1.24, with any vaccine currently in use has been elusive.
95% CI 0.89–1.72). Risk of developing multiple sclerosis The purpose of this study was to systematically review
remained unchanged after BCG, Hepatitis B, Influenza, all randomized clinical trials (RCTs) and non-randomized
studies (NRS) reporting on risk of developing MS, or a
relapse, following any immunization recommended by the
Electronic supplementary material The online version of this CDC for use in children, adults, or travelers, and also BCG.
article (doi:10.1007/s00415-011-5984-2) contains supplementary
material, which is available to authorized users. The need of a systematic review on the role of vacci-
nations in MS providing doctors and patients with a clear
M. F. Farez
J. Correale (&) reference source on the subject was the main objective of
Department of Neurology, Dr Raúl Carrea Institute
this paper, as well as to encourage future well-designed
for Neurological Research, FLENI, Montañeses 2325,
1428 Buenos Aires, Argentina studies investigating how specific immunizations affect MS
e-mail: jcorreale@fleni.org.ar; bairesla@fibertel.com.ar patients in particular.

123
1198 J Neurol (2011) 258:1197–1206

Methods variability between studies, values of 40% or higher were

considered significant and no pooled estimate was
A detailed method protocol was specified and documented calculated.
in advance by both authors. This review was prepared Large studies are usually published regardless of effect
following the PRISMA statement guidelines [5] for sys- size, but small studies have to be significant to be pub-
tematic reviews and meta-analyses (Preferred Reporting lishable, thus leading to a potential ‘‘publication bias’’. We
Items for Systematic Reviews and Meta-Analyses) (see assessed this possibility by performing Begg and Egger
Supplementary Material for the PRISMA checklist). Even tests. All analyses were performed using STATA v11.1
though we expected the number of NRS to be greater than (Statacorp LP, TX, USA).
the number of RCTs, we chose PRISMA guidelines over
MOOSE guidelines (Meta-Analysis of Observational
Studies in Epidemiology) because we considered them to Results
be more updated and better suited to both NRS and RCTs.
A detailed description of the methods is provided as sup- Study selection
plementary material (Appendix 1).
We systematically searched MEDLINE (1966–January A total of 1,342 studies were identified through database
2011), EMBASE (1977–January 2011) and the Cochrane searching of Medline, EMBASE and CENTRAL. Seven
Central Register of Controlled Trials (CENTRAL) (1961– additional studies were identified through reference mining
January 2011) for randomized controlled trials and non- of review articles and relevant publications. Of these, 287
randomized studies examining the effect of immunizations duplicates were removed and a total of 1,062 reports
in the risk of MS and the risk of relapse in adult MS, in screened. After reviewing the abstracts, 932 studies were
English and in other languages. In addition, we hand discarded because they did not clearly meet the inclusion
searched abstracts of selected conferences from 2005 to criteria. Full text of the remaining 130 studies was assessed
2011, including AAN, EFNS, ENS, and ECTRIMS. Ref- for eligibility, and 107 articles were excluded. Thus, 23
erences from relevant studies and previous reviews were studies were included in the qualitative synthesis and 14 in
also searched. the quantitative synthesis. The selection process is shown
Non-randomized studies and randomized clinical trials in the flow diagram in Fig. 1, and studies included are
written in English or in other languages, describing the summarized in Supplementary Tables 1 and 2. The results
effect of immunizations on risk of developing MS or risk of of risk of bias assessment within and across studies are
disease relapse were included. Case reports and case report summarized for each individual study in the Supplemen-
series were excluded. All immunizations recommended by tary Table 3.
the CDC [6] for use in children, adults or travelers were We found at least one study in 18 of the original 23
analyzed, and BCG was added to the list. vaccines examined. Enough information was gathered to
In duplicate and independently the two authors screened obtain a pooled estimate for eight of these vaccines.
all the studies and selected the articles that satisfied the
eligibility criteria. The data was extracted onto a stan- Study characteristics
dardized data extraction sheet by one of the authors (MFF)
and checked independently by the second author (JC). BCG
Disagreements were solved by checking the articles, con-
tacting authors and by consensus when needed. Six studies were included in the analysis [9–14]. All were
RCTs and NRS were analyzed separately for each vac- NRS, two nested case–control and the remaining four case–
cine. Risk ratio (RR) was used for RCTs and cohort stud- control studies. The outcome measured was the odds of
ies, and odds ratio (OR) for other study designs. developing MS. We did not find any evidence of a publi-
For vaccines with three or more studies published, cation reporting risk of relapse in MS patients. The studies
results of individual studies were statistically combined to included involved 536 MS cases and 751 controls selected
obtain a pooled estimate of the ‘‘average’’ effect on risk of from the general population, the same hospitals or from the
relapse or MS. This procedure was only applied if studies participating neurology department.
did not show evidence of heterogeneity (i.e., significant In all six studies, vaccination status was assessed ret-
variability between studies not due to chance). For the rospectively, with only one report [9] using health records
pooled estimate we used the DerSimonian and Laird ran- to ascertain status. The rest used questionnaires, either self-
dom effects model with inverse variance weights, and for administered, or performed by interviewers.
heterogeneity we used the Cochran’s Q statistic and the I2 As summarized in Fig. 2a and Supplementary Table 1,
statistic [7, 8]. I2 statistic measures the proportion of all studies showed a consistent null effect of BCG on risk

123
J Neurol (2011) 258:1197–1206 1199

Fig. 1 Selection of studies for

inclusion in the systematic
review on the effect of
immunizations on risk of
developing multiple sclerosis
and risk of MS relapse. A total
of 23 reports were included in
the qualitative synthesis and 14
in the final meta-analysis

of developing MS (OR 0.96, 95% CI 0.69–1.34) with no study showing data regarding the risk of relapse. One study
evidence of heterogeneity as measured by the I2 statistic was a nested case–control and the remaining two were
(0.0% and p value of 0.950). No evidence of publication case–control reports. A total of 237 MS cases and 387
bias, or bias of any other source was observed, and the controls were included, and all used questionnaires to
average Newcastle–Ottawa (NOS) scale rating was 5.5 (see assess immunization.
Supplementary Table 3). Overall, evidence found suggests As shown in Fig. 2b, pooled OR showed decreased risk
BCG immunization may not be associated with increased of developing MS with diphtheria immunization, (0.60,
risk of developing MS. 95% CI 0.40–0.90 and p value of 0.014). Even though we
found no evidence of heterogeneity (I2 = 0.0%
Chickenpox (Varicella) p = 0.627) and we thus calculated a pooled estimate, all
the studies showed high risk of bias in the domain of
We could find only one study linking risk of developing blinding of participants, personnel and outcome assessors.
MS to chickenpox immunization, including 140 patients This fact, together with the small number of patients
and 131 controls [14], in which authors reported a signif- included, should be taken into account when interpreting
icantly high OR for chickenpox immunization and MS (OR these results. Thus, there is some evidence that diphtheria
41.6, 95% CI 5.6–309.6). However, significance of these immunization may be associated with a decreased risk of
results was limited to a univariate analysis; no association developing MS.
was found in multivariate analysis, including socio-eco-
nomic, environmental and family related variables. Thus, Diphtheria, pertussis, tetanus (DPT)
there is insufficient evidence on chickenpox immunization
and the risk of developing MS or relapse. We could find only one study on the risk of developing MS
with DPT immunization [10]. The study, which included
Diphtheria 38 MS patients and 46 controls (relatives), reported an OR
for DPT immunization and MS of 0.8 with a 95% CI
We found three studies reporting on the risk of developing 0.29–2.11. Thus, there is insufficient evidence to support
MS after diphtheria immunization [12, 13, 15], and no any role of DPT immunization in MS.

123
1200 J Neurol (2011) 258:1197–1206

Fig. 2 Study specific and

pooled odds ratio/relative risks
from studies on different
immunization and MS. CI
confidence interval

123
J Neurol (2011) 258:1197–1206 1201

Hepatitis A 0.873. We found little evidence of heterogeneity

(I2 = 5.1% p = 0.368). There is, therefore, suggestive
We found one study on risk of developing MS after Hep- evidence against an increased risk of developing MS by
atitis A immunization [16]. The study by DeStefano et al. influenza immunization (Fig. 2d)
[16] included 440 MS patients and 950 controls, and Pooled relative risk (RR) of relapse following influenza
reported an OR for Hepatitis A vaccination and MS of 2.17 immunization was 1.24 with a 95% CI of 0.89–1.72 and a
(with a 95% CI 0.5–9.5). We found no study showing data p value of 0.2. We found no evidence of heterogeneity
regarding the risk of relapse. Thus, there is insufficient (I2 = 0.0% p = 0.531) (Fig. 2e). Hence, the evidence
evidence to support any role of Hepatitis A immunization found suggests influenza immunization is not associated
on risk of developing MS. with increased risk of relapse.
We could not find any published evidence on effects of
Hepatitis B the new H1N1 vaccine and MS, probably due to its recent
market release.
We found eight NRS that addressed risk of developing MS
[4, 16–22], and one study reporting risk of relapse [23]. Of Measles
the former, four were nested case–control studies, two were
case–control studies, one was a cohort study, and one was a We found seven NRS addressing risk of MS [13–16, 20,
before-and-after comparison. A total of 50,117 MS cases 28, 29], and we were unable to find studies reporting the
and 140,333 controls were included in the analysis of case– risk of relapse. A total of 15,372 MS cases and 9,969
control studies. controls were included in the analysis of case–control
Overall, the majority assessed immunization status studies.
consulting health records, avoiding potential recall bias. When results of individual studies were pooled in a
See Supplementary Table 3 for detailed results of risk of meta-analysis, we obtained an OR of 1.164 with a 95% CI
bias and quality assessment of each study. of 0.754–1.798, but found evidence of significant hetero-
As summarized in Supplementary Table 1 and Fig. 2c, a geneity (I2 = 70.6%, p = 0.002). We were only able to
pooled estimate of the effect of the six case–control studies obtain a pooled estimate without significant heterogeneity
was calculated, obtaining a null OR of 1.00 (95% CI (I2 = 4.1%, p = 0.390) when the study by Zorzon et al.
0.74–1.37). Although p value was not significant was removed (OR 1.074, 95% CI of 0.974–1.185). Poten-
(p = 0.099), evidence of moderate heterogeneity was tial explanations for the discrepancy between Zorzon et al.
found, with an I2 of 47.7%. We, therefore, explored potential and the other studies may lie in the fact that controls
sources of heterogeneity and found that the study by Geier chosen were not sampled from the general population, but
and Geier [4] compared HBV immunization against tetanus from blood donors, which may be different from MS
immunization, while the others did not. When this study was patients or the general population. In addition, vaccination
excluded, heterogeneity was eliminated I2 of 0.00% assessment was conducted by un-blinded face to face
(p = 0.870) and a more precise OR of 0.92 with 95% CI interview, thus potentially introducing recall bias (MS
0.84–1.004 was obtained. Regarding risk of relapse, the patients were more aware of their health, or interviewers
case-crossover study by Confavreux et al. [23] reported a asked more detailed questions, thus increasing chances of
relative relapse risk of 0.67 with a 95% CI 0.2–2.17. remembering being vaccinated). The evidence found sug-
Thus, there is suggestive evidence that Hepatitis B gests measles immunization is not associated with changes
immunization is not associated with an increased risk of in risk of developing MS.
developing MS, and there is insufficient evidence in the
case of risk of relapse. Measles, mumps, rubella (MMR)

Influenza We found three NRS that addressed the risk of MS [10, 16,
28], and we were unable to find studies reporting the risk of
We found four studies describing OR of developing MS MS relapse. A total of 676 MS cases and 1,879 controls
after influenza immunization [14, 16, 18, 20] and five were included in the meta-analysis of case–control studies.
studies showing risk of relapse [23–27]. A total of 14,997 Pooled OR of developing MS and MMR immunization
cases, and 10,128 controls were included in the meta- was 1.02 with a 95% CI of 0.64–1.61 and a p value of
analysis on risk of developing MS; and 156 cases and 157 0.704 (see Fig. 2f). We found no evidence of heterogeneity
controls were analyzed for risk of relapse. (I2 = 0.0%, p = 0.622). Hence, there is suggestive evi-
Pooled OR of developing MS and influenza immuniza- dence that risk of MS is not altered by previous MMR
tion was 0.97 with a 95% CI of 0.77–1.23 and a p value of immunization.

123
1202 J Neurol (2011) 258:1197–1206

Mumps Rabies

We found five NRS addressing risk of developing MS We found only two studies that addressed the risk of
[13–15, 20, 28], and we were unable to find studies developing MS with rabies vaccination [10, 12] and both
reporting risk of MS relapse. A total of 14,841 MS cases reports show a null association.
and 8,927 controls were included in the meta-analysis of Thus, there is insufficient evidence to support a role of
case–control studies. Rabies immunization in MS.
Pooled OR of developing MS and mumps immunization
was 2.71 with a 95% CI of 0.81–9.09 but we found evi- Rubella
dence of considerable heterogeneity (I2 = 73.8%,
p = 0.004). We were able to obtain a pooled estimate We found five NRS that addressed risk of MS [12, 14, 16,
without significant heterogeneity (I2 = 0.00%, p = 0.763) 20, 28], and were unable to find studies reporting the risk of
once again, when the study by Zorzon et al. was removed relapse. A total of 15,252 MS cases and 9,790 controls
(OR 1.09, 95% CI 1.02–1.18). Therefore, the evidence were included in the meta-analysis.
suggests mumps immunization is not associated with Pooled OR of developing MS and rubella immunization
changes in risk of developing MS. was 1.11 with a 95% CI of 0.65–1.90 and a p value of
0.704. We found evidence of high heterogeneity
Pertussis (I2 = 84.6%, p = 0.0001); this time, the study of Zorzon
et al. could not explain heterogeneity as it persisted despite
We found two NRS that addressed risk of MS [13, 15], and removal of the study from the analysis. Hence, there is
we were unable to find studies reporting risk of MS relapse. conflicting evidence regarding rubella immunization and
The study by Kurtzke et al. [15] included 23 MS patients risk of developing MS.
and 127 controls, and reported an OR for MS of 0.41 (95%
CI 0.04–1.88). Work by Pekmezovic et al. [13] included Tetanus
110 MS patients and 110 controls and reported also
absence of MS risk (OR 1.0, 95% CI 0.6–1.8). Thus, there We found eight NRS addressing risk of MS [12, 13, 15, 16,
is insufficient evidence to support any role of Pertussis 18, 29–31], and one study reporting the risk of MS relapse
immunization in MS. [23]. A total of 5,526 MS cases and 1,680 controls were
included in the meta-analysis of case–control studies for
Pneumococcal risk of developing MS.
Pooled OR of developing MS and tetanus immunization
We were able to find only one report of the risk of was 0.68 with a 95% CI of 0.54–0.84 and a p value of
developing MS and anti-pneumococcal immunization and 0.0001. We found little evidence of heterogeneity
no evidence regarding the risk of MS relapse [16]. The (I2 = 8.3%, p = 0.366) (Fig. 2h). Despite the lack of
study by DeStefano et al. found no association (OR of 1.20 heterogeneity, and even though the number of cases and
and 95% CI of 0.31–4.02), but, as suggested by the authors, controls was higher than for diphtheria (the other immu-
too few cases or controls had been vaccinated in order to nization potentially showing decreased MS risk), caution
obtain an accurate estimate. Thus, there is insufficient should be used when interpreting these results. As shown in
evidence to support an association between anti-pneumo- Supplementary Table 3, some of the studies included in the
coccal immunization and risk of MS. meta-analysis have NOS scores below 5, and high risk of
bias in more than one domain. Therefore, evidence sug-
Polio gests that tetanus immunization may be associated with
decreased risk of developing MS, but further research is
We found seven NRS that reported risk of MS [10–15, 29], needed to confirm this finding.
and we were unable to find studies reporting risk of MS
relapse. A total of 570 MS cases and 725 controls were Typhoid fever
included in the meta-analysis.
Pooled OR of developing MS and polio immunization We found four NRS addressing risk of developing MS
showed a null association with an OR of 0.87 and a 95% CI [12, 15, 29, 30], and were unable to find studies reporting the
of 0.61–1.25 (p = 0.462). We found evidence of some risk of MS relapse. A total of 393 MS cases and 529 controls
heterogeneity (I2 = 28.2%) but p value was not significant were included in the meta-analysis of case–control studies.
(p = 0.213) (see Fig. 2g). Thus, polio immunization may We found no association between MS and typhoid fever
not be associated with an increased risk of developing MS. immunization with an OR of 1.05 and a 95% CI of

123
J Neurol (2011) 258:1197–1206 1203

0.72–1.53 and a p value of 0.704. We found little but not unclear. In addition to the caveats mentioned before, we
significant evidence of heterogeneity (I2 = 14.5%, should add that for studies included in the diphtheria meta-
p = 0.320) (Fig. 2i). Therefore, the evidence suggests analysis, it remains unclear whether any distinction was
there may not be any change in the risk of developing MS made between DPT and diphtheria alone; these results
after typhoid fever immunization. should therefore be taken with caution.
There is consensus that certain criteria should be met to
Yellow fever determine causality, (i.e. tetanus immunization is the cause
of the decreased risk of developing MS). In this case, since
We found two reports on yellow fever immunization and immunization preceded disease onset by many years and
risk of developing MS [12, 15]. The study by Casetta et al. there is no evidence of latent infection or the lack of
[12], including 104 MS patients, 150 controls with an OR infections, per se, to meet causality criteria, as defined by
of developing MS of 0.35 (95% CI 0.00–3.64), and the Bradford Hill [44], is difficult.
report by Kurtzke et al. [15], including 23 MS patients, 127 We are fully aware of the limitations of a meta-analysis
controls and an OR of 0.73 (95% CI 0.01–6.10). Thus, of mostly observational studies, and that interpretation of
there is insufficient evidence to support any role of yellow overall results should be considered with caution. In
fever immunization on risk of developing MS. addition, combining data across studies to obtain pooled
estimates is also subject to the drawback caused by the fact
that MS patient ethnic characteristics, type of immuniza-
Discussion tion and other variables may not be the same across the
different studies. However, due to the nature of the inter-
All the evidence reported in this meta-analysis is summa- vention, and as a result of ethical and practical issues, most
rized in Table 1, including quality of the evidence grading of the data found came from prospective or case–control
using the GRADE system [32]. studies, as will probably be the case for future data gen-
MS is a demyelinating disorder, the precise etiology of erated on the topic too, for that matter. With the generation
which still remains largely unclear [33], yet it is commonly of reporting guidelines for observational studies [45], more
agreed that the onset of disease is influenced by genetic and standardized studies could be performed in the future and
environmental factors [34]. The importance of the envi- more power obtained to ascertain potential associations
ronment has been recently highlighted by a study showing between vaccines and MS.
lack of genetic differences between monozygotic twins In addition, some studies were separated by as much as
discordant for MS [35]. Evidence for involvement of the 20 years, with the obvious changes in vaccine manufac-
environment in MS development comes from a variety of turing technologies as well as serotypes. For example, in
epidemiological studies linking MS to latitude [36], vita- the case of influenza and risk of relapse, four studies were
min D levels [37], smoking [38], and exposure to certain identified which included four different serotypes, three
infections (mainly EBV) [39]. In the case of infections, different manufacturers and 20 years elapsed between
many studies show the association of bacterial and viral some of them [24–27]. To address if the pooled estimates
infections with the onset of relapses [40, 41], while we of the vaccines that were included in the meta-analysis
have previously shown beneficial influence of chronic have shifted over time, we performed a cumulative meta-
parasite infections on disease course [42]. It would, analysis [46], where the evidence at the time a new study
therefore, be valid to consider that immunization against a was published is calculated. We did not find any significant
pathogen associated with higher risk of MS or relapse, changes over time for any of the vaccines (data not shown).
would likely diminish the risk of disease. On the other Unfortunately, we did not have sufficient data to stratify
hand, immunizations have been reported in some studies to our analysis for other factors, and specific effects of strains
be associated with increased risk of MS and relapse. These and vaccine components remains elusive.
observations, together with case reports and some media Evidence linking immunization to relapses is scarce.
coverage, may lead to doctors and patients to have doubts A study by Confavreux et al. [23] following a case-cross-
over immunizations given these conflicting results. This over design, reports no association for Hepatitis B or teta-
meta-analysis shows that none of the immunizations nus. Likewise, for influenza, there is also limited
included in the quantitative analysis were associated with information. Even though we did not include case-reports or
higher risk of developing MS or higher risk of relapse. In case-series in our study, we found a considerable number
fact, diphtheria and tetanus immunization may be associ- describing demyelinating events at variable time periods
ated with a lower risk of MS, in agreement with a previ- after several immunizations [47–52]. These findings may be
ously published report [43]. The mechanism through which conflictive for both doctors and patients when deciding on
these two vaccines may exert a protective effect remains immunization, especially in the case of vaccines that do not

123
1204 J Neurol (2011) 258:1197–1206

Table 1 Summary of evidence

Vaccine Risk of Quality of Risk of a Quality of References
developing MS evidence MS relapse evidence
for risk of for risk of MS
developing relapse
MS (GRADE) (GRADE)

BCG Null Moderate No data NA [9–14]

OR (95% CI) 0.96
(0.69–1.34)
Chickenpox Unclear NA No data NA [14]
Diphtheria Decreased Very low No data NA [12, 13, 15]
OR (95% CI) 0.6 (0.4–0.9)
DPT Unclear NA No data NA [10]
Hib No data NA No data NA NA
Hepatitis A Unclear NA No data NA [16]
Hepatitis B Null Moderate Unclear NA [4, 16–23]
OR (95% CI) 1.00
(0.74–1.37)
HPV No data NA No data NA NA
Influenza Null Low Null Moderate [14, 16, 18, 20, 23–27]
OR (95% CI) 0.97 OR (95% CI) 1.24
(0.77–1.23) (0.89–1.72)
Japanese No data NA No data NA NA
encephalitis
Measles Unclear NA No data NA [13–16, 20, 28, 29]
Meningococcal No data NA No data NA NA
MMR Null Low No data NA [10, 16, 28]
OR (95% CI) 0.93
(0.62–1.38)
Mumps Unclear NA No data NA [13–15, 20, 28]
Pertussis Unclear NA No data NA [15]
Pneumococcal Unclear NA No data NA [16]
Polio Null Moderate No data NA [10–15, 29]
OR (95% CI) 0.87
(0.61–1.25)
Rabies Unclear NA No data NA [10]
Rotavirus No data NA No data NA NA
Rubella Unclear NA No data NA [12, 14, 16, 28]
Tetanus Decreased Low Unclear NA [12, 13, 15, 16, 18, 23, 29–
OR (95% CI) 0.68 31]
(0.54–0.84)
Typhoid Fever Null Low No data NA [12, 15, 23, 29, 30]
OR (95% CI) 1.05
(0.72–1.53)
Yellow Fever Unclear NA No data NA [15, 23]
NA not applicable

form part of mandatory national immunization programs or influenza, MMR, polio, tetanus and typhoid fever vaccines.
are recommended for travelers. Well-designed studies are We also report that diphtheria and tetanus immunization
necessary to address these concerns. may be associated with a decreased risk of developing MS,
In conclusion, the review reveals no increase in risk of and conclude further research is needed for varicella, DPT,
developing MS with BCG, hepatitis B, diphtheria, haemophilus influenza b, hepatitis A, HPV, Japanese

123
J Neurol (2011) 258:1197–1206 1205

encephalitis, Measles, anti-meningococcal, mumps, per- Multiple sclerosis in the Faroe Islands 7 Results of a case control
tussis, rabies, rotavirus, rubella, and yellow fever to questionnaire with multiple controls. Acta Neurol Scand
96(3):149–157
ascertain potential association. 16. DeStefano F, Verstraeten T, Jackson LA, Okoro CA, Benson P,
Black SB, Shinefield HR, Mullooly JP, Likosky W, Chen RT,
Acknowledgments This study was supported by funding from the Vaccine Safety Datalink Research Group NIP, Centers for Dis-
Raúl Carrea Institute for Neurological Research, FLENI. ease Control, Prevention (2003) Vaccinations and risk of central
nervous system demyelinating diseases in adults. Arch Neurol
Conflict of interest None. 60(4):504–509. doi:10.1001/archneur.60.4.504
17. Ascherio A, Zhang SM, Hernán MA, Olek MJ, Coplan PM,
Brodovicz K, Walker AM (2001) Hepatitis B vaccination and the
risk of multiple sclerosis. N Engl J Med 344(5):327–332
18. Hernán MA, Jick SS, Olek MJ, Jick H (2004) Recombinant
hepatitis B vaccine and the risk of multiple sclerosis: a pro-
References spective study. Neurology 63(5):838–842
19. Hocine MN, Farrington CP, Touzé E, Whitaker HJ, Fourrier A,
1. WHO (2005) Immunization against diseases of public health Moreau T, Tubert-Bitter P (2007) Hepatitis B vaccination and
importance. World Health Organization (WHO). http://www.who. first central nervous system demyelinating events: reanalysis of a
int/mediacentre/factsheets/fs288/en/index.html. Accessed 25 Aug case-control study using the self-controlled case series method.
2010 Vaccine 25(31):5938–5943. doi:10.1016/j.vaccine.2007.05.032
2. Doja A, Roberts W (2006) Immunizations and autism: a review 20. Ramagopalan SV, Valdar W, Dyment DA, DeLuca GC, Yee IM,
of the literature. Can J Neurol Sci 33(4):341–346 Giovannoni G, Ebers GC, Sadovnick AD, Group CCS (2009)
3. Chen RT, Pless R, Destefano F (2001) Epidemiology of auto- Association of infectious mononucleosis with multiple sclerosis.
immune reactions induced by vaccination. J Autoimmun A population-based study. Neuroepidemiology 32(4):257–262.
16(3):309–318. doi:10.1006/jaut.2000.0491 doi:10.1159/000201564
4. Geier DA, Geier MR (2005) A case–control study of serious 21. Touzé E, Fourrier A, Rue-Fenouche C, Rondé-Oustau V, Jean-
autoimmune adverse events following hepatitis B immunization. taud I, Bégaud B, Alpérovitch A (2002) Hepatitis B vaccination
Autoimmunity 38(4):295–301 and first central nervous system demyelinating event: a case–
5. Liberati A, Altman DG, Tetzlaff J, Mulrow C, Gøtzsche PC, control study. Neuroepidemiology 21(4):180–186
Ioannidis JPA, Clarke M, Devereaux PJ, Kleijnen J, Moher D 22. Zipp F, Weil JG, Einhäupl KM (1999) No increase in demye-
(2009) The PRISMA statement for reporting systematic reviews linating diseases after hepatitis B vaccination. Nat Med
and meta-analyses of studies that evaluate healthcare interven- 5(9):964–965. doi:10.1038/12376
tions: explanation and elaboration. BMJ 339:b2700 23. Confavreux C, Suissa S, Saddier P, Bourdès V, Vukusic S,
6. CDC (2010) Immunization schedule. Center for Disease Control ViMSS Group (2001) Vaccinations and the risk of relapse in
and Prevention of the United States of America. http://www.cdc. multiple sclerosis. Vaccines in Multiple Sclerosis Study Group.
gov/vaccines/recs/schedules/default.htm. Accessed 22 Aug 2010 N Engl J Med 344(5):319–326
7. Higgins JP, Thompson SG, Deeks JJ, Altman DG (2003) Mea- 24. Bamford CR, Sibley WA, Laguna JF (1978) Swine influenza
suring inconsistency in meta-analyses. BMJ 327(7414):557–560. vaccination in patients with multiple sclerosis. Arch Neurol
doi:10.1136/bmj.327.7414.557 35(4):242–243
8. Takkouche B, Cadarso-Suarez C, Spiegelman D (1999) Evalua- 25. Miller AE, Morgante LA, Buchwald LY, Nutile SM, Coyle PK,
tion of old and new tests of heterogeneity in epidemiologic meta- Krupp LB, Doscher CA, Lublin FD, Knobler RL, Trantas F,
analysis. Am J Epidemiol 150(2):206–215 Kelley L, Smith CR, La Rocca N, Lopez S (1997) A multicenter,
9. Andersen E, Isager H, Hyllested K (1981) Risk factors in multiple randomized, double-blind, placebo-controlled trial of influenza
sclerosis: tuberculin reactivity, age at measles infection, tonsil- immunization in multiple sclerosis. Neurology 48(2):312–314
lectomy and appendectomy. Acta Neurol Scand 63(2):131–135 26. Mokhtarian F, Shirazian D, Morgante L, Miller A, Grob D,
10. Bansil S, Singhal BS, Ahuja GK, Riise T, Ladiwala U, Behari M, Lichstein E (1997) Influenza virus vaccination of patients with
Cook SD (1997) Multiple sclerosis in India: a case–control study multiple sclerosis. Multiple Scler 3(4):243–247
of environmental exposures. Acta Neurol Scand 95(2):90–95 27. Myers LW, Ellison GW, Lucia M, Novom S (1976) Swine-
11. Berr C, Puel J, Clanet M, Ruidavets JB, Mas JL, Alperovitch A influenza vaccination in multiple sclerosis. N Engl J Med
(1989) Risk factors in multiple sclerosis: a population-based case- 295(21):1204
control study in Hautes-Pyrénées, France. Acta Neurol Scand 28. Ahlgren C, Torén K, Odén A, Andersen O (2009) A population-
80(1):46–50 based case–control study on viral infections and vaccinations and
12. Casetta I, Granieri E, Malagù S, Tola MR, Paolino E, Caniatti subsequent multiple sclerosis risk. Eur J Epidemiol
LM, Govoni V, Monetti VC, Fainardi E (1994) Environmental 24(9):541–552. doi:10.1007/s10654-009-9367-2
risk factors and multiple sclerosis: a community-based, case- 29. Zilber N, Kahana E (1996) Risk factors for multiple sclerosis: a
control study in the province of Ferrara, Italy. Neuroepidemiol- case–control study in Israel. Acta Neurol Scand 94(6):395–403
ogy 13(3):120–128 30. Compston DA, Vakarelis BN, Paul E, McDonald WI, Batchelor
13. Pekmezovic T, Jarebinski M, Drulovic J (2004) Childhood infec- JR, Mims CA (1986) Viral infection in patients with multiple
tions as risk factors for multiple sclerosis: Belgrade case–control sclerosis and HLA-DR matched controls. Brain 109(Pt
study. Neuroepidemiology 23(6):285–288. doi:10.1159/000080094 2):325–344
14. Zorzon M, Zivadinov R, Nasuelli D, Dolfini P, Bosco A, Bratina 31. Alter M, Speer J (1968) Clinical evaluation of possible etiologic
A, Tommasi MA, Locatelli L, Cazzato G (2003) Risk factors of factors in multiple sclerosis. Neurology 18(2):109–116
multiple sclerosis: a case–control study. Neurol Sci 32. Atkins D, Best D, Briss PA, Eccles M, Falck-Ytter Y, Flottorp S,
24(4):242–247. doi:10.1007/s10072-003-0147-6 Guyatt GH, Harbour RT, Haugh MC, Henry D, Hill S, Jaeschke
15. Kurtzke JF, Hyllested K, Arbuckle JD, Brønnum-Hansen H, R, Leng G, Liberati A, Magrini N, Mason J, Middleton P, Mru-
Wallin MT, Heltberg A, Jacobsen H, Olsen A, Eriksen LS (1997) kowicz J, O’Connell D, Oxman AD, Phillips B, Schünemann HJ,

123
1206 J Neurol (2011) 258:1197–1206

Edejer TT-T, Varonen H, Vist GE, Williams JW, Zaza S, Group GW 42. Correale J, Farez M (2007) Association between parasite infec-
(2004) Grading quality of evidence and strength of recommenda- tion and immune responses in multiple sclerosis. Ann Neurol
tions. BMJ 328(7454):1490. doi:10.1136/bmj.328.7454.1490 61(2):97–108. doi:10.1002/ana.21067
33. Weiner HL (2009) The challenge of multiple sclerosis: how do 43. Hernán MA, Alonso A, Hernández-Dı́az S (2006) Tetanus vacci-
we cure a chronic heterogeneous disease? Ann Neurol nation and risk of multiple sclerosis: a systematic review. Neurology
65(3):239–248. doi:10.1002/ana.21640 67(2):212–215. doi:10.1212/01.wnl.0000225079.51201.f9
34. Ramagopalan SV, Dobson R, Meier UC, Giovannoni G (2010) 44. Hill AB (1965) The environment and disease: association or
Multiple sclerosis: risk factors, prodromes, and potential causal causation? Proc R Soc Med 58:295–300
pathways. Lancet Neurol 9(7):727–739. doi:10.1016/S1474-4422 45. Vandenbroucke JP, von Elm E, Altman DG, Gotzsche PC,
(10)70094-6 Mulrow CD, Pocock SJ, Poole C, Schlesselman JJ, Egger M
35. Baranzini SE, Mudge J, van Velkinburgh JC, Khankhanian P, (2007) Strengthening the Reporting of Observational Studies in
Khrebtukova I, Miller NA, Zhang L, Farmer AD, Bell CJ, Kim RW, Epidemiology (STROBE): explanation and elaboration. PLoS
May GD, Woodward JE, Caillier SJ, McElroy JP, Gomez R, Pando Med 4(10):e297
MJ, Clendenen LE, Ganusova EE, Schilkey FD, Ramaraj T, Khan 46. Lau J, Antman EM, Jimenez-Silva J, Kupelnick B, Mosteller F,
OA, Huntley JJ, Luo S, Kwok P-Y, Wu TD, Schroth GP, Oksenberg Chalmers TC (1992) Cumulative meta-analysis of therapeutic
JR, Hauser SL, Kingsmore SF (2010) Genome, epigenome and trials for myocardial infarction. N Engl J Med 327(4):248–254.
RNA sequences of monozygotic twins discordant for multiple doi:10.1056/NEJM199207233270406
sclerosis. Nature 464(7293):1351–1356. doi:10.1038/nature 47. Cabrera-Gómez JA, Echazabal-Santana N, Garcı́a-González L,
08990 Ramos-Cedeño AM, Rodrı́guez-Roque MO, López-Hernández O,
36. Hernán MA, Olek MJ, Ascherio A (1999) Geographic variation Cabrera-Núñez JA, De González La Nuez J, Téllez L (2002) A
of MS incidence in two prospective studies of US women. severe episode in a patient with recurrent disseminated acute
Neurology 53(8):1711–1718 encephalitis due to vaccination against hepatitis B. For or against
37. Munger KL, Zhang SM, O’Reilly E, Hernán MA, Olek MJ, vaccination? Rev Neurol 34(4):358–363
Willett WC, Ascherio A (2004) Vitamin D intake and incidence 48. Chaves M, Riccio P, Patrucco L, Rojas J, Cristiano E (2009)
of multiple sclerosis. Neurology 62(1):60–65 Longitudinal myelitis associated with yellow fever vaccination.
38. Hernán MA, Jick SS, Logroscino G, Olek MJ, Ascherio A, Jick H J Neurovirol 15(4):348–350. doi:10.1080/13550280903062805
(2005) Cigarette smoking and the progression of multiple scle- 49. Miller H, Cendrowski W, Shapira K (1967) Multiple sclerosis
rosis. Brain 128(Pt 6):1461–1465. doi:10.1093/brain/awh471 and vaccination. Br Med J 2(5546):210–213
39. Ascherio A, Munger KL (2007) Environmental risk factors for 50. Sutton I, Lahoria R, Tan I, Clouston P, Barnett M (2009) CNS
multiple sclerosis. Part I. The role of infection. Ann Neurol demyelination and quadrivalent HPV vaccination. Multiple Scler
61(4):288–299. doi:10.1002/ana.21117 15(1):116–119. doi:10.1177/1352458508096868
40. Correale J, Fiol M, Gilmore W (2006) The risk of relapses in 51. Terney D, Beniczky S, Barsi P, Kondákor I, Perényi J, Faludi B,
multiple sclerosis during systemic infections. Neurology Szapper M, Vécsei L (2006) Multiple sclerosis after hepatitis B
67(4):652–659. doi:10.1212/01.wnl.0000233834.09743.3b vaccination in a 16-year-old patient. Chin Med J 119(1):77–79
41. Buljevac D, Flach HZ, Hop WCJ, Hijdra D, Laman JD, Savelkoul 52. Wildemann B, Jarius S, Hartmann M, Regula JU, Hametner C
HFJ, van Der Meché FGA, van Doorn PA, Hintzen RQ (2002) (2009) Acute disseminated encephalomyelitis following vacci-
Prospective study on the relationship between infections and nation against human papilloma virus. Neurology 72(24):2132–
multiple sclerosis exacerbations. Brain 125(Pt 5):952–960 2133. doi:10.1212/WNL.0b013e3181aa53bb

123