G Model

JVAC 16538 1–14 ARTICLE IN PRESS

Vaccine xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

Vaccine
journal homepage: www.elsevier.com/locate/vaccine

1 Review

2 Risk of febrile seizure after measles–mumps–rubella–varicella

3 vaccine: A systematic review and meta-analysis
4 Q1 Shu-Juan Ma, Yi-Quan Xiong, Li-Na Jiang, Qing Chen ∗
5 Department of Epidemiology, School of Public Health and Tropical Medicine, Southern Medical University, 1838 Guangzhou North Road,
6 Guangzhou 510515, China
7

8
22 a r t i c l e i n f o a b s t r a c t
9
10 Article history: Background: Considering the febrile seizure rate, there is no longer a clear preference for use of
11 Received 2 March 2015 measles–mumps–rubella–varicella (MMRV) vaccine over separate measles–mumps–rubella (MMR) and
12 Received in revised form 31 May 2015 varicella (V) vaccine. This work was undertaken to assess the risk of febrile seizure after MMRV vaccine
13 Accepted 1 June 2015
in children.
14 Available online xxx
Methods: We searched PubMed, Embase, BIOSIS Previews, Scopus, Web of Science, Cochrane Library and
15
other databases through 12 December 2014. Meta-analysis was conducted using R version 3.1.2 and Stata
16 Keywords:
version 12.0.
17 Measles–mumps–rubella–varicella vaccine
18 Febrile seizure
Results: A total of thirty-nine studies were included. Thirty-one published or unpublished clinical trials
19 Randomized controlled trial involving about 40,000 subjects did not show significant differences in incidence of febrile seizure or
20 Post-marketing observation vaccine related febrile seizure between MMRV and MMR with or without varicella vaccine after any
21 Meta-analysis doses, in the risk windows of 0–28, 0–42 or 0–56 days and 7–10 days. In addition, these studies showed
that the receipt of concomitant use of MMRV and other pediatric vaccine was not a significant predictor of
febrile seizure. Eight post-marketing observations involving more than 3,200,000 subjects were included.
No evidence suggested elevated risk of febrile seizure associated with MMRV vaccine among children
aged 4–6 years old during 7–10 days or 0–42 days after vaccination. However, an approximately 2-fold
increase in risk of seizure or febrile seizure during 7–10 days or 5–12 days after MMRV vaccination was
found among children aged 10–24 months, although the highest incidence of seizure was still lower than
2.95‰.
Conclusions: First MMRV vaccine dose in children aged 10–24 months was associated with an elevated
risk of seizure or febrile seizure. Further post-marketing restudies based on more rigorous study design
are needed to confirm the findings.
© 2015 Published by Elsevier Ltd.

23 1. Introduction Pre-licensure clinical trials have demonstrated that MMRV 33

vaccines are generally well tolerated and have comparable 34

24Q2 Combination measles–mumps–rubella–varicella (MMRV) vac- immunogenicity and overall safety profiles to MMR or MMR + V 35

25 cine was originally designed as an alternative to separate vaccines in healthy children, except higher fever rate following the 36

26 measles–mumps–rubella (MMR) and varicella (V) vaccines, based administration of MMRV vaccine [6–9]. Fever can precipitate febrile 37

27 on similar vaccination schedules and good concomitant safety seizure in susceptible children aged between 6 months and 5 years, 38

28 profiles [1–3]. Two MMRV vaccines have been available since mid- especially in the range of 12–3 months [10,11]. 39

29 2000s with different formulations: ProQuad (Merck&Co., Inc, West Febrile seizure is the most common neurologic adverse event 40

30 Point, PA) and Priorix-Tetra (GlaxoSmithKline Biologicals, Rixen- following immunization with measles-containing vaccines [12]. 41

31 sart, Belgium). Both were developed based on the existing MMR Given the evidence for an increase in risk of febrile seizure associ- 42

32 and varicella vaccines [4,5]. ated with the use of combination vaccine, the Advisory Committee 43

on Immunization Practices has no longer expressed a preference 44

for use of MMRV vaccine over MMR + V vaccine [13]. And since 45

then, febrile seizure following MMRV vaccine has been investi- 46

∗ Corresponding author. Tel.: +86 20 61648312; fax: +86 20 61648312. gated intensively. Although febrile seizures are common and not 47

E-mail address: [email protected] (Q. Chen). associated with long-term neurologic sequelae or developmental 48

http://dx.doi.org/10.1016/j.vaccine.2015.06.009
0264-410X/© 2015 Published by Elsevier Ltd.

Please cite this article in press as: Ma S-J, et al. Risk of febrile seizure after measles–mumps–rubella–varicella vaccine: A systematic
review and meta-analysis. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.06.009
G Model
JVAC 16538 1–14 ARTICLE IN PRESS
2 S.-J. Ma et al. / Vaccine xxx (2015) xxx–xxx

49 delay, they often frighten parents, which may precipitate acute 2.5. Quality assessment 105

50 care visits and undermine confidence in immunization programs

51 [14,15]. For this reason, the incidence and timing of vaccine related The quality of included clinical trials was assessed using the 106

52 febrile seizure have to be carefully studied [16]. Jadad score [17], applying a score from 0 (poor quality) to 5 (rig- 107

53 Up to date, a number of studies focused on the incidence orous), based on randomization and adequate performance of the 108

54 of febrile seizure following MMRV vaccine have been published. randomization procedure, double-blind and the adequate perfor- 109

55 However, the risks of febrile seizure following different measles- mance, description of withdrawals and dropouts. Additionally, the 110

56 containing vaccines have not been directly compared, or discrepant quality of observational studies was assessed using the Newcastle- 111

57 among studies. To achieve a comprehensive evidence-based sug- Ottawa Scale (NOS) [18,19]. In this scale, studies are scored across 112

58 gestion for the fair use of MMRV vaccine, we conducted a systematic three categories: selection of subjects (four stars), comparability 113

59 review and meta-analysis of clinical trials and post-marketing of study groups (two stars), and assessment of outcome/exposure 114

60 observational safety surveillance studies to evaluate the risk of (three stars). The star rating system was used to indicate the quality 115

61 febrile seizure after MMRV vaccine in children. of a study, with a maximum of nine stars. Studies were graded on an 116

ordinal star scoring scale with higher scores representing studies 117

62 2. Methods of higher quality. 118

63 2.1. Eligibility criteria 2.6. Statistical analysis 119

64 Eligible study designs were randomized controlled trials (RCTs), Risk ratios (RRs) or risk differences (RDs) with binomial 95% 120
65 quasi-RCTs, controlled clinical trials, uncontrolled clinical trials, confidence intervals (CIs) for binary results were calculated. 121
66 cohort studies, case control studies or cross-sectional studies, Between-study heterogeneity was assessed by using Cochrane Q 122
67 involving MMRV, MMR, MMR + V vaccines and other pediatric statistic and quantified by estimated I2 . A Mantel–Haenszel fixed- 123
68 vaccines co-administered with MMRV vaccine. The population of effects model (M–H, fixed) was used to calculate when the test 124
69 interest were children aged 0–6 years, irrespective of sex and eth- for heterogeneity was not statistically significant (P > 0.10), other- 125
70 nic origin. The numbers or incidence of febrile seizure or vaccine wise the DerSimonian–Laird random-effects model (DL, random) 126
71 related febrile seizure had been reported. was employed [20–22]. In meta-analysis, all statistical tests were 127

two-sided, possible significant difference was indicated if 95% CI 128

72 2.2. Literature search of RD excluded 0 and 95% CI of RR excluded 1. Meta-analysis was 129

considered to be inappropriate when the I2 statistic exceeded 50% 130

73 We searched PubMed, Embase, BIOSIS Previews (1994–2013), [23]. Between-study heterogeneity was explored by MMRV vaccine 131

74 Scopus, Web of Science, Cochrane Library, National Insti- manufacturer, comparison category and vaccination age. We also 132

75 tutes of Health database (clinicaltrials.gov), GSK Clinical performed sensitivity analysis to evaluate whether any single study 133

76 Trials (gsk-clinicalstudyregister.com), MERCK Clinical Trials dominated the results of the meta-analyses. Finally, publication 134

77 (merck.com/clinical-trials) from the earliest date available bias was assessed by visual inspection of funnel plots for outcomes 135

78 through 12 December 2014. We used key words or subject head- reported by more than five studies [24], and then quantificationally 136

79 ings for (“measles” and “mumps” and “rubella” and “varicella”) or detected using Begg test and Egger test [24,25]. Statistical analy- 137

80 (“measles mumps rubella varicella” or “MMRV”) in combination ses were conducted using R version 3.1.2 and Stata version 12.0 138

81 with “vaccine” and “seizure”, including their common synonyms. (StataCorp). 139

82 There were no language restrictions. We screened bibliographies of

83 selected original studies, review articles and relevant conference
3. Results 140
84 abstracts, and contacted corresponding authors of studies for
85 missing data.
3.1. Description of studies included 141

86 2.3. Study selection The searches identified 532 potentially relevant studies and clin- 142

ical trial records. Most ineligible studies were excluded on the basis 143
87 Citations were merged together in Endnote, version X7 to of information in the title or abstract. The selection process is shown 144
88 facilitate management. Two reviewers independently applied the in Fig. 1. 145
89 inclusion criteria to all retrieved articles in an un-blinded standard- We included 31 clinical trials (30 RCTs) in the review (Table 1) 146
90 ized manner, evaluated by title, abstract and full text successively. [26–50]. Twenty-three clinical trials [28–50] were published and 147
91 Disagreements were resolved by consensus. the others [26,27] were unpublished. Twenty-nine clinical trials 148

[26–35,37–50] reported the incidence of febrile seizure or vaccine 149

92 2.4. Data collection related febrile seizure within 0–28, 0–42 or 0–56 days (0–28/42/56 150

days) after first MMRV vaccine dose in healthy children aged 151
93 Data were mainly obtained from published articles. Unpub- 9–24 months, twenty-five of them [26–30,33–35,37–43,45–50] 152
94 lished clinical trials served as a supplement. Priority was given to also reported the incidence within a smaller risk window of 7 to 153
95 published data in case of subtle discrepancy. The more complete or 10 days (7–10 days) after vaccination. Children in 17 clinical tri- 154
96 recent publications were given precedence if there were multiple als [27–30,33,35,37,38,40,41,44–46,48,50] received two doses of 155
97 publications from the same study or data source. For each eligi- MMRV vaccines. Jadad scores were equivalent to 2 or 3 in all the 156
98 ble study, the following data elements were selectively extracted clinical trials listed in Table 1, except the study by Rüger, et al. 157
99 onto piloted structured forms independently by two reviewers: (1) [44], which was not a RCT and was not included in the quan- 158
100 general study data: first author, publication year, country, study ID, titative meta-analysis. Considering little difference between the 159
101 study design, data source; (2) study population: total subjects, age, quality of included RCTs, there was not a stratified analysis by 160
102 gender; (3) treatment: MMRV vaccine type, vaccine group, num- ranking. 161
103 ber of doses, duration of follow up; (4) outcome: number of febrile Eight post-marketing observational studies for evaluating the 162
104 seizure cases and vaccine related febrile seizure cases, adjustment. adverse events of MMRV vaccine were also included in this review 163

Please cite this article in press as: Ma S-J, et al. Risk of febrile seizure after measles–mumps–rubella–varicella vaccine: A systematic
review and meta-analysis. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.06.009
G Model
JVAC 16538 1–14 ARTICLE IN PRESS
S.-J. Ma et al. / Vaccine xxx (2015) xxx–xxx 3

Records idenfied through database searching (N =445): Addional records idenfied through other
Idenficaon
PubMed (n=29); Embase (n=175); sources (N = 87): clinicaltrials.gov (n=34);

BIOSIS Previews (n=16); Scopus (n=162); GSK clinical trials (n=23);

Web of Science (n=51); Cochrane Library (n=12). Merck clinical trials (n=5); Bibliographies (n=25).

Records excluded (N=153):

Titles and abstracts screened aer Ineligible arcle type, e.g. review (n=93)
duplicates removed (N=246) No MMRV vaccinaon (n=47)
Screening

No relevant outcomes (n=13)

Records excluded (N=54):

Duplicates (n=8)

Ineligible arcle type, e.g. review (n=10)

Full-text arcles assessed for eligibility
Methodology (n=6)
Eligibility

(N =93)
No extractable data (n=10)

No MMRV vaccinaon (n=15)

No relevant outcomes (n=5)

Studies included in review (N=39):

Included

Clinical trials (n=31)

Cohort (n=7)

Self-controlled case series (n=1)

Fig. 1. PRISMA flow diagram of study selection. GSK, GlaxoSmithKline Biologicals; Merck, Merck & Co.; MMRV, measles–mumps–rubella–varicella vaccine.

164 (Table 3) [12,51–57], seven of which were retrospective cohort definition, the latter was defined as closely as possible as the crite- 186

165 studies [12,51–54,56,57] and the other one was self-controlled case ria specified by Jacobsen et al. [51]; (4) chart-confirmed according 187

166 series study [55]. All cohort studies included had more than 8 stars to several chart reviews [52]; (5) retrieved using some coded 188

167 assessed using NOS. Two observational studies were from the same and free-text diagnoses [54]; (6) accepted as a case a physician 189

168 data source [12,52], and only the recent one was included in the diagnosis [53]. 190

169 pooled analysis [12].

3.3. Clinical trials 191

170 3.2. Definition of febrile seizure

Thirty-one clinical trials involving about 40,000 subjects 192

171 Febrile seizure in two published [44,50] and all unpublished reported incidence of febrile seizure or vaccine related febrile 193

172 clinical trials was reported as one of serious adverse events (SAEs), seizure (Table 1). During 0–28/42/56 days after first vaccine dose, a 194

173 while it was not limited to SAE in other published trials. Vaccine total of 71 febrile seizure cases were reported in 31,949 vaccinated 195

174 related febrile seizure was defined as febrile seizure diagnosed by children aged 9–24 months from 26 RCTs [26–35,37–43,46,48–50], 196

175 the investigator to have a causal relationship to investigational vac- and 21 vaccine related febrile seizure cases were reported in 197

176 cine. A seizure event was defined as the first instance during the 42 34,292 vaccinated children aged 9–24 months from 26 clinical 198

177 days after MMRV vaccination, by using The International Classi- trials [26–31,33–35,37–41,43–50]. Within a smaller risk window 199

178 fication of Diseases, 9th Revision (ICD-9) codes 345*(epilepsy) or of 7–10 days after first vaccine dose, 11 febrile seizure cases 200

179 780.3*(convulsion) in the emergency department or hospital in and 9 vaccine related febrile seizure cases were reported respec- 201

180 4 studies [12,52,53,55]. In study by MacDonald, et al. [56], only tively in 13,129 and 25,433 vaccinated children aged 9–24 months 202

181 ICD-9 code 780.3*(convulsion) was used for convulsions. Identi- [26–30,33–35,37–43,45–50]. In 16 studies informed the prognosis 203

182 fication of febrile seizure was one in the following list: (1) ICD, [27–37,39,40,45,48,49], a total of 42 febrile seizures were reported, 204

183 10th revision, Canadian version, codes R56.0* (febrile convulsion) and these cases all recovered without any sequelae. 205

184 [56]; (2) criteria for convulsion developed by the Brighton Collab- There were five following kinds of vaccine groups involved: 206

185 oration [51,58]; (3) “FC narrow” [57], a subset of “FC Jacobsen” by (1) MMRV group, children received MMRV vaccine alone; (2) 207

Please cite this article in press as: Ma S-J, et al. Risk of febrile seizure after measles–mumps–rubella–varicella vaccine: A systematic
review and meta-analysis. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.06.009
4

JVAC 16538 1–14

G Model
review and meta-analysis. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.06.009
Please cite this article in press as: Ma S-J, et al. Risk of febrile seizure after measles–mumps–rubella–varicella vaccine: A systematic

Table 1
Characteristics of included published/unpublished clinical trials.

Study Study ID Country Study Jadad MMRV Follow-up Age at dose No. of Time First MMRV dose Second MMRV dose
design score vaccine period one (mean doses between
type age) doses

Vaccine group Total FS case Vaccine Vaccine Total FS case Vaccine

subject related FS group subject related FS

GSK208136/ GSK208136/ Germany, RCT 2 Priorix- 0–42 d 12–24 m 1 NA MMRV 370 1 0 [0]
006 2001 [26] 006 Estonia, Tetra (15.0 m)
Belgium
MMR + V 186 0 0 [0]

GSK208136/ GSK208136/ Australia, RCT 2 Priorix- 0–42 d 12–24 m 1 NA MMRV 1538 2 0 [0]
007 2001 [26] 007 Canada, Tetra (14.1 m)
Italy,
Colombia,

ARTICLE IN PRESS
Mexico
MMR + V 767 1 0 [0]
2 6w MMR 763 0 0 [0] V 763 0 [0] 0 [0]

S.-J. Ma et al. / Vaccine xxx (2015) xxx–xxx

GSK208136/ GSK208136/ Australia, RCT 2 Priorix- 0–42 d 12–24 m 1 NA MMRV 486 2 0 [0]
016 2001 [26] 016 Belgium Tetra (12.8 m)
MMR 66 1 0 [0]

NCT00126997 GSK104020/ Finland, RCT 2 Priorix- 0–42 d 11–21 m 2 6w MMRV 1225 AD:1 0 [0] MMRV 1205 AD:1 0 [0]
2005 [27] NCT00126997 Germany, Tetra (14.0 m)
Greece,
Poland
MMR + V 213 0 0 [0] MMR 207 0 0 [0]

Shinefield NCT00975507 America RCT 3 ProQuad 0–42 d 12–23 m 2 3m MMRV 323 1 [1] 1 [1] MMRV 310 0 [0] 0 [0]
(1)2005 [28] (14.1 m)
1 NA MMR + V 157 0 [0] 0 [0]

Shinefield NCT00986232 America RCT 3 ProQuad 0–42 d 11–23 m 2 3m MMRVb 764 2 [2] 1 [1] MMRVb 719 0 [0] 0 [0]
(2)2005 [29] (12.9 m)
1 NA MMR + V 381 0 [0] 0 [0]

Knuf, 2006 [30] GSK208136/ Germany, RCT 2 Priorix- 0–42 d 12–24 m 2 6–8 w MMRV 371 1 [1] 0 [0] MMRV 364 1 [1] 1 [1]
038 Austria Tetra (14.4 m)
MMR + V 123 0 [0] 0 [0] MMR 122 0 [0] 0 [0]

Lieberman, NCT00985153 America, RCT 2 ProQuad 0–42 d 11–23 m 1 NA MMRV 2860 8 4

2006 [31] Canada (12.7 m)
MMR + V 993 3 1

Shinefield, NCT00984295 America RCT 2 ProQuad 0–56 d 11–16 m 1 NA MMRV + Hib/ 929 2
2006 [32] (12.4 m) HepB + DTaP
MMRV 479 1
MMR + V 467 3

NCT00326183 NCT00326183 America RCT 2 ProQuad 0–28 d 12–17 m 2 24 w HAV + MMRV 347 1 [0] 0 [0] HAV + MMRV 347 0 [0] 0 [0]
2007 [27] (13.1 m)
HAV 1453 0 [0] 0 [0] HAV 1453 0 [0] 0 [0]

Goh, 2007 [33] GSK208136/ Singapore RCT 2 Priorix- 0–42 d 9–10 m 2 3m MMRV 151 0 [0] 0 [0] MMRV 138 0 [0] 0 [0]
018 Tetra (9.3 m)
MMR + V 144 0 [0] 0 [0] MMR + V 136 1 [0] 1 [0]
JVAC 16538 1–14
G Model
review and meta-analysis. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.06.009
Please cite this article in press as: Ma S-J, et al. Risk of febrile seizure after measles–mumps–rubella–varicella vaccine: A systematic

Table 1 (Continued)

Study Study ID Country Study Jadad MMRV Follow-up Age at dose No. of Time First MMRV dose Second MMRV dose
design score vaccine period one (mean doses between
type age) doses

Vaccine group Total FS case Vaccine Vaccine Total FS case Vaccine

subject related FS group subject related FS

Zepp, 2007 [34] GSK208136/ Germany RCT 2 Priorix- 0–42 d 12–23 m 1 NA MMRV + DTaP- 150 0 [0] 0 [0]
013 Tetra (13.7 m) HBV-IPV/Hib
MMRV 150 1 [1] 0 [0]
DTaP-HBV- 150 0 [0] 0 [0]
IPV/Hib

NCT00373958 NCT00373958 America RCT 3 ProQuad 0–30 d 12–15 m 1 NA PCV- 267 0

2008 [27] 13 + MMRV +
Pedvax-
HIB + HAV

ARTICLE IN PRESS
PCV-7 + MMRV 258 1
+ Pedvax-
HIB + HAV

S.-J. Ma et al. / Vaccine xxx (2015) xxx–xxx

Schuster, 2008 GSK104389/ Germany RCT 2 Priorix- 0–42 d 10–21 m 2 6w MMRV 732 AD:3 1 [1] MMRV 725 AD:3 0 [0]
[35] NCT00127023 Tetra (12.9 m)
MMR + V 238 AD:1 0 [0] MMR 236 AD:1 0 [0]

Halperin, 2009 GSK105909/ Italy, RCT 3 Priorix- 0–42 d 15 m-6y 1 ≥6 w MMR + V MMRV 195 0 [0] 0 [0]
a
[36] NCT00352898 Canada Tetra (47.0 m)
MMR + V 194 0 [0] 0 [0]

NCT00422292 NCT00422292 America RCT 2 ProQuad 0–6 m 12 m 1 NA Menactra 257 3

2009 [27]
Menactra + 616 6
MMRV
Menactra + PCV 246 1
MMRV + PCV 476 4

Knuf, 2010 [37] GSK106670/ Germany RCT 2 Priorix- 0–42 d 11–21 m 2 6w MMRV(SC) 162 0 [0] 0 [0] MMRV (SC) 161 0 [0] 0 [0]
NCT00351923 Tetra (12.6 m)
MMRV(IM) 166 1 [1] 0 [0] MMRV (IM) 161 1 [0] 0 [0]

Vesikari, 2010 GSK107706/ Finland RCT 3 Priorix- 0–42 d 12–14 m 2 6-8 w MMRV + PHiD- 109 0 [0] 0 [0] MMRV + 108 0 0
[38] NCT00370227 Tetra (12.3 m) CV DTaP-HBV-
IPV/Hib
2 6-8 w MMRV + DTaP- 101 AD:1 [0] 0 [0] MMRV + 99 AD:1 0 [0]
HBV-IPV/Hib PHiD-CV [0]
1 NA PHiD-CV + 112 0 [0] 0 [0]
DTaP-HBV-
IPV/Hib

Leonardi, 2011 NCT00109343 America RCT 3 ProQuad 0–28 d 12–15 m 1 NA MMRV + PCV-7 510 1 [1] 0 [0]
[39] (12.6 m)
2 6w PCV-7 258 0 [0] 0 [0] MMRV 246 0 [0] 0 [0]
2 6w MMRV 259 1 [0] 0 [0] PCV-7 246 0 [0] 0 [0]

NCT00969436 GSK109995/ India RCT 2 Priorix- 0–42 d 9–10 m 2 6m MMRV 180 0 [0] 0 [0] MMRV 180 0 [0] 0 [0]
2011 [27] NCT00969436 Tetra (9.0 m)
MMR 180 AD:1 0 [0] MMRV 180 AD:1 0 [0]
MMR 90 0 [0] 0 [0] MMR + V 90 0 [0] 0 [0]

5
6

JVAC 16538 1–14

G Model
review and meta-analysis. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.06.009
Please cite this article in press as: Ma S-J, et al. Risk of febrile seizure after measles–mumps–rubella–varicella vaccine: A systematic

Table 1 (Continued)

Study Study ID Country Study Jadad MMRV Follow-up Age at dose No. of Time First MMRV dose Second MMRV dose
design score vaccine period one (mean doses between
type age) doses

Vaccine group Total FS case Vaccine Vaccine Total FS case Vaccine

subject related FS group subject related FS

Rümke, 2011 GSK103388/ Germany, RCT 2 Priorix- 0–42 d 11–13 m 2 4w MMRV 183 0 [0] 0 [0] MMRV 181 0 [0] 0 [0]
[40] NCT00127010 Belgium, Tetra (12.0 m)
Netherlands
12 m MMRV 179 1 [0] 0 [0] MMRV 155 0 [0] 0 [0]

ARTICLE IN PRESS
4w MMR 185 1 [0] 0 [0] MMR 183 1 [0] 0 [0]

Vesikari, 2011 GSK109670/ Finland RCT 2 Priorix- 0–42 d 12–23 m 2 12 w ACWY- 375 1 [0] 0 [0] MMRV 375 0 [0] 0 [0]
[41] NCT00474266 Tetra (14.6 m) TT + MMRV

S.-J. Ma et al. / Vaccine xxx (2015) xxx–xxx

6w ACWY-TT 374 0 [0] 0 [0] MMRV 374 0 [0] 0 [0]
MMRV 126 0 [0] 0 [0] MenC- 126 0 [0] 0 [0]
CRM197
MenC-CRM197 125 1 [0] 0 [0] MMRV 125 0 [0] 0 [0]

Blatter, 2012 GSK110058/ America RCT 2 Priorix- 0–42 d 12–15 m 1 NA MMRV 676 2 [0]
[42] NCT00578175 Tetra/ (12.3 m) (GSK + 4 ◦ C) +
ProQuad PCV-7 + HAV
MMRV(GSK- 668 2 [0]
20 ◦ C) + PCV-
7 + HAV
MMRV(Merck- 379 1 [0]
20 ◦ C) + PCV-
7 + HAV

Klein, 2012 NCT00626327 America RCT 2 ProQuad 0–28 d 12 m 1 NA MenACWY- 504 0 [0] 0 [0]
[43] CRM + MMRVc
MenACWY- 510 0 [0] 0 [0]
CRM
MMRVc 616 1 [1] 0 [0]

Rüger, 2012 NCT00560755 Europe No RCT 0 ProQuad 0–28 d 12–22 m 2 4-6 w MMRV(rHA) 3376 7 MMRV 3342 4
[44] (14.6 m)

Vesikari, 2012 NCT00566527 Finland, RCT 2 ProQuad 0–28 d 9 m(9.5 m) 2 90-104d MMRV(rHA) 528 0 [0] MMRV(rHA) 524 0 [0]
[45] France,
Germany
11 m(11.3 m) MMRV(rHA) 480 0 [0] MMRV(rHA) 474 0 [0]
12 m(12.3 m) MMRV(rHA) 528 0 [0] MMRV(rHA) 462 0 [0]
JVAC 16538 1–14
G Model
review and meta-analysis. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.06.009
Please cite this article in press as: Ma S-J, et al. Risk of febrile seizure after measles–mumps–rubella–varicella vaccine: A systematic

Table 1 (Continued)

Study Study ID Country Study Jadad MMRV Follow-up Age at dose No. of Time First MMRV dose Second MMRV dose
design score vaccine period one (mean doses between
type age) doses

Vaccine group Total FS case Vaccine Vaccine Total FS case Vaccine

subject related FS group subject related FS

Huang, 2013 GSK108760/ Singapore, RCT 3 Priorix- 0–42 d 11–22 m 2 12 w MMRV(new 330 2 [0] 0 [0] MMRV 327 1 [0] 1 [0]
[46] NCT00892775 Taiwan Tetra (13.0 m) WS) (new WS)
MMRV 166 1 [0] 0 [0] MMRV 164 0 [0] 0 [0]

ARTICLE IN PRESS
Vesikari, 2013 NCT00657709/ Europe RCT 3 Priorix- 0–28 d 12 m 1 NA 4CMenB + 765 1 [1]
[47] 00847145 Tetra MMRV
2 1m 4CMenB 789 0 [0] MMRV 789 0 [0]

S.-J. Ma et al. / Vaccine xxx (2015) xxx–xxx

Yetman, 2013 NCT00312858 America RCT 2 ProQuad 0–28 d 12–16 m 2 24 w HAV + MMRV + 330 1 [1] 0 [0] HAV + MMRV 330 1 [0] 0 [0]
[48] (12.4 m) PCV-7
34 w MMRV + PCV-7 323 1 [1] 1 [1] MMRV 323 1 [0] 0 [0]

Cha, 2014 [49] GSK110876/ Korea RCT 3 Priorix- 0–42 d 11–23 m 1 NA MMRV 310 2 [1] 0 [0]
NCT00751348 Tetra (12.0 m)
MMR + V 159 0 [0] 0 [0]

Prymula, 2014 GSK100388/ Europe RCT 3 Priorix- 0–42 d 11–22 m 2 6w MMRV 2489 AD/G:13 3 [3] MMRV 2489 AD/G:13 0 [0]
[50] NCT00226499 Tetra (14.2 m)
MMR 2487 0 [0] V 2487 0 [0]
MMR 827 1 [1] MMR 827 0 [0]

Note: Data from result summaries of clinical trials opened on the websites were used as a supplement. AD, across doses; AD/G, across doses/groups; d, days; FS, febrile seizure; GSK, GlaxoSmithKline Biologicals; m, months;
MMRV, measles–mumps–rubella–varicella vaccine; NA, not applicate; RCT, randomized controlled trial; w, weeks; y, years.
Study, author and published year represented published study, GSK or/and ClinicalTrials.gov identifier and the year finished represented unpublished study.
MMRV vaccine type: ProQuad, MMRV vaccine manufactured by Merck & Co; Priorix-Tetra, MMRV vaccine manufactured by GlaxoSmithKline.
Vaccine group: 4CMenB, multicomponent, recombinant, meningococcal serogroup B vaccine; ACWY-TT, meningococcal ACWY-tetanus toxoid conjugate vaccine; DTaP, diphtheria, tetanus and acellular pertussis vaccine;
DTPa-HBV-IPV/Hib, Diphtheria-tetanus-acellular pertussis-hepatitisB-inactivated poliovirus-Haemophilus influenzae type b conjugate vaccine; HAV, hepatitis A vaccine; Hib/HepB, combined Haemophilus influenzae type b
conjugate-hepatitis B vaccines; Menactra, Meningococcal (Groups A, C, Y and W-135) Polysaccharide Diphtheria Toxoid Conjugate Vaccine; MenACWY-CRM, quadrivalent meningococcal conjugate vaccine; MenC-CRM197,
meningococcal serogroup C conjugate vaccine; MMR, measles–mumps–rubella vaccine; MMRV(IM), MMRV vaccine administrated intramuscularly; MMRV(GSK + 4 ◦ C), Priorix-Tetra stored at 4 ◦ C; MMRV(GSK-20 ◦ C), Priorix-Tetra
stored at −20 ◦ C; MMRV(Merck −20 ◦ C), ProQuad stored at −20 ◦ C; MMRV(new WS): MMRV vaccine manufactured with measles and rubella monovalent bulks derived from a newly established working seed virus stock;
MMRV(rHA), MMRV vaccine manufactured with recombinant human albumin(rHA) instead of human serum albumin(HSA); MMRV(SC), MMRV vaccine administrated subcutaneously; PCV-7, pneumococcal 7-valent conjugate
vaccine; PCV-13, pneumococcal 13-valent conjugate vaccine; PHiD-CV, 10-valent pneumococcal nontypeable Haemophilus influenzae protein D conjugate vaccine; PedvaxHib, Haemophilus b conjugate vaccine; V, varicella
vaccine.
a
, age at second MMR-related dose of children previously immunized with MMR and varicella vaccines.
b
, only the middle and high doses met criteria for the minimum clinically acceptable varicella virus dose were included.
c
a subset of subjects (16% and 29%, respectively) received separate MMR plus varicella rather than ProQuad owing to a shortage of ProQuad in 2009.
[n]: number of cases reported in the time window of 7–10 days after vaccination.

7
G Model
JVAC 16538 1–14 ARTICLE IN PRESS
8 S.-J. Ma et al. / Vaccine xxx (2015) xxx–xxx

208 MMR + V group, children received MMR and varicella vaccines Additionally, four clinical trials studied the influence of admin- 274

209 simultaneously at two separate anatomic sites; (3) MMR group, istration route, composition and vaccination age on the safety of 275

210 children received MMR vaccine alone; (4) MMRV + Others group, MMRV vaccine [37,44–46]. There was no statistical difference in the 276

211 children received MMRV vaccine and one or more other pedi- incidence of febrile seizure or vaccine related febrile seizure among 277

212 atric vaccines concomitantly, like DTaP (diphtheria, tetanus and different administration routes, compositions and vaccination ages 278

213 acellular pertussis vaccine), Hib/HepB (combined Haemophilus of MMRV vaccine. 279

214 influenzae type b conjugate-hepatitis B vaccines), ACWY-TT

215 (meningococcal ACWY-tetanus toxoid conjugate vaccine), PCV-
3.4. Post-marketing observational studies 280
216 7 (pneumococcal 7-valent conjugate vaccine) and so on. (5)
217 Others group, children received one or more other pediatric vac-
Post-marketing evaluation of MMRV vaccine included seven ret- 281
218 cines. Considering different comparison groups in the clinical
rospective cohort studies [12,51–54,56,57] and one self-controlled 282
219 trials, hierarchical analyses of four comparisons were employed.
case series [55]. Details are presented in Table 3. RRs for single study 283
220 Summarized results are listed in Table 2. No evidence of hetero-
and pooled RRs are listed in Table 4. No heterogeneity was found 284
221 geneity was found in any comparison (I2 = 0%). Finer stratifications
in all the appropriate meta-analyses (I2 ranged from 0% to 4.9%). 285
222 by MMRV vaccine type and vaccination age were not carried
Four retrospective cohort studies observed seizures. During 286
223 out.
7–10 days after vaccination, the incidence ranges in children aged 287
224 Twenty-one RCTs [26–35,39–41,43,47,49,50] were included in
12–23 months were 0.93 to 1.29‰, 0.46 to 0.54‰, and 0.29 to 0.56‰ 288
225 the meta-analysis to analyze the incidence of febrile seizure
for MMRV, MMR + V and MMR groups, respectively [12,52,56]. 289
226 and vaccine related febrile seizure within 0–28/42/56 days after
Pooled seizure risk was significantly higher in MMRV group than 290
227 first MMRV vaccine dose in healthy children aged 9–24 months
MMR + V (RR 2.00; 95% CI 1.66, 2.40) and MMR (RR 1.90; 95% CI 291
228 (Table 2), nineteen of them [26–30,33–35,39–41,43,47,49,50] com-
1.43, 2.53) groups. The self-controlled case series study suggested 292
229 pared the incidence in a risk window of 7–10 days after vaccination.
that receipt of MMRV vaccine between 16 and 23 months was asso- 293
230 During 0–28/42/56 days after first MMRV dose, pooled incidence
ciated with more post-vaccination seizures than between 12 and 294
231 of febrile seizure was under 2.51‰ in all groups. The incidence
15 months, receipt of MMRV vaccine compared with MMR vaccine 295
232 of febrile seizure was numerically higher in MMRV group than
doubled the RR for post-vaccination seizures during 7–10 days after 296
233 MMR + V group (2.51‰ vs. 2.07‰) and MMR group (1.95‰ vs. 0‰),
vaccination [55]. 297
234 so was the incidence of vaccine related febrile seizure (0.81‰ vs.
During 0–42 days after vaccination, the incidence ranges 298
235 0.30‰ and 0.59‰ vs. 0.22‰). In the risk window of 7–10 days after
of seizure in children aged 12–23 months were 2.27–2.95‰, 299
236 first MMRV dose, pooled incidence of febrile seizure and vaccine
1.59–2.10‰ and 1.04–1.92‰ in MMRV, MMR + V and MMR groups, 300
237 related febrile seizure was also numerically higher in MMRV group
respectively [12,52,56]. The risk was also significantly higher in 301
238 than MMR + V group (2.61‰ vs. 0‰ and 0.52‰ vs. 0‰), and the
MMRV vaccine recipients, compared with MMR + V (RR 1.27; 95% 302
239 incidence in MMRV group was no less than that in MMR group.
CI 1.41, 1.42) and MMR (RR 1.40; 95% CI 1.19, 1.65) vaccine recipi- 303
240 However, the analysis of pooled RDs and RRs suggested that, there
ents. Nevertheless, incidence of seizure in children aged 4–6 years 304
241 were no significant differences in the incidence of febrile seizure or
was much lower in all the groups (around 0.20‰ during 0–42 days 305
242 vaccine related febrile seizure in MMRV vs. MMR + V and MMRV vs.
and around 0.02‰ during 7–10 days) [53]. No significant increase 306
243 MMR comparisons, in the risk windows of 0–42/56 days and 7–10
of seizure was found either 7–10 days or 0–42 days after adminis- 307
244 days first MMRV dose.
tering MMRV vaccine compared with MMR + V or MMR vaccines in 308
245 Incidence of febrile seizures and vaccine related febrile seizures
children aged 4–6 years. 309
246 was similar or lower within 0–28/42 days or 7–10 days after
During the second week after immunization, the incidence 310
247 second MMRV vaccine dose than that after first MMRV vac-
ranges of febrile seizure in children aged 10–24 months were 311
248 cine dose in most studies [27–29,35,37,40,44–46,48,50]. Ten RCTs
0.62–0.96‰, 0.32–0.44‰ and 0.25–0.42‰ in MMRV, MMR + V and 312
249 [27,30,33,35,36,40,48,50] were included in the meta-analysis to
MMR groups, respectively [51,52,54,57]. An elevated risk of febrile 313
250 analyze the incidence of febrile seizure and vaccine related febrile
seizure was demonstrated in MMRV group, both compared with 314
251 seizure within 0–28/42 days or 7–10 days after second MMRV vac-
MMR + V vaccination group (RR was 2.04 during 7–10 days and 315
252 cine dose in healthy children aged 9–24 months (Table 2), one
1.77 during 5–12 days) and MMR vaccination group (RR was 2.36 316
253 or even zero febrile seizure or vaccine related febrile seizure was
during 7–10 days and 2.32 during 5–12 days). During 0–30/40/42 317
254 reported after the second dose in each group in these studies
days after vaccination, the incidence ranges of febrile seizure 318
255 (Table 1). No statistical differences in the incidence of febrile seizure
in children aged 10–24 months were 1.20–2.28‰, 0.62–1.28‰ 319
256 or vaccine related febrile seizure were found in any of the four com-
and 0.84–2.19‰ in MMRV, MMR + V and MMR groups, respec- 320
257 parisons in Table 2, in the risk windows of 0–28/42 days and 7–10
tively [51,52,54,57]. Significant increase of febrile seizure during 321
258 days after second MMRV vaccine dose.
0–30/40/42 days after immunization with MMRV vaccine was only 322
259 Twelve RCTs [27,32,34,38,39,41–43,47] studied the safety of
found when it was compared with MMR + V (RR 1.37; 95% CI 1.12, 323
260 concomitant use of MMRV vaccine with some other pediatric
1.69) vaccines. However, this result was not found in children aged 324
261 vaccines at similar administration schedule. Seven of them were
4–6 years (RR 2.33; 95% CI 0.10, 57.25), which only one febrile 325
262 included in the meta-analysis [27,32,34,39,41,43,47]. Ignoring dif-
seizure diagnosis was confirmed during 7–10 days post MMRV 326
263 ferent concomitant pediatric vaccines, during 0–28/42/56 days and
vaccination [53]. 327
264 7–10 days after vaccination, incidence of febrile seizure was numer-
265 ically highest in MMRV group, followed by MMRV + Others group,
266 and lowest in Others group, only one vaccine related febrile seizure 3.5. Publication bias 328

267 was found in MMRV + Others group in the two comparisons. There
268 were no statistical differences in the incidence of febrile seizure Potential publication bias was found in the analyses of pooled 329

269 or vaccine related febrile seizure among MMRV + Others group RRs of vaccine related febrile seizure in MMRV vs. MMR + V compar- 330

270 involved comparisons. Similar results were found in another five ison, by using Begg test (P = 0.003 for the risk window of 0–42/56 331

271 RCTs [27,38,42,48], in which MMRV vaccines were co-administered days, P = 0.012 for the risk window of 7–10 days) and Egger test 332

272 with one or more other pediatric vaccines in all comparison groups (P = 0.018 for the risk window of 0–42/56 days, P = 0.011 for the 333

273 (Table 1). risk window of 7–10 days). 334

Please cite this article in press as: Ma S-J, et al. Risk of febrile seizure after measles–mumps–rubella–varicella vaccine: A systematic
review and meta-analysis. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.06.009
JVAC 16538 1–14
G Model
review and meta-analysis. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.06.009
Please cite this article in press as: Ma S-J, et al. Risk of febrile seizure after measles–mumps–rubella–varicella vaccine: A systematic

Table 2
Pool analyses of incidence of febrile seizure and vaccine related febrile seizure following vaccination in healthy children aged 9–24 months.

Dose Comparison categories (risk Febrile seizure Vaccine related febrile seizure
window/group)

No. of Pooled Pooled RD Pooled RR No. of Pooled Pooled RD Pooled RR

studies incidence (‰, 95% CI) (95% CI) studies incidence (‰, 95% CI) (95% CI)
(‰) (‰)

First MMRV dose MMRV vs. MMR + V

0–42/56 d MMRV 9 2.51 (18/7166) 0.47 (−1.71, 2.66) 1.02 (0.47, 2.23) 10 0.81 (7/8644) 0.50 (−1.10, 2.10) 0.87 (0.31, 2.40)
MMR + V 2.07 (7/3377) 0.30 (1/3361)
7–10 d MMRV 5 2.61 (5/1919) 2.50 (−1.77, 6.78) 1.50 (0.35, 6.38) 9 0.52 (3/5784) 0.56 (−1.50, 2.61) 0.74 (0.23, 2.36)
MMR + V 0 (0/964) 0 (0/2368)
MMRV vs. MMR
0–42 d MMRV 4 1.95 (5/2566) −0.40 (−4.20, 3.40) 0.71 (0.19, 2.74) 5 0.59 (3/5055) 0.58 (−0.78, 1.94) 1.39 (0.39, 4.98)
MMR 1.81 (2/1104) 0.22 (1/4598)

ARTICLE IN PRESS
7–10 d MMRV 2 0 (0/542) 0 (−7.98, 7.98) 0.51 (0.03, 8.08) 5 0.59 (3/5055) 0.58 (−0.78, 1.94) 1.39 (0.39, 4.98)
MMR 0 (0/275) 0.22 (1/4598)
MMRV vs. MMRV + others
0–28/42/56 d MMRV 5 2.45 (4/1630) 1.07 (−2.27, 4.42) 1.63 (0.46, 5.74) 4 0 (0/1151) 0 (−3.31, 3.31) 1.44 (0.22, 9.61)

S.-J. Ma et al. / Vaccine xxx (2015) xxx–xxx

MMRV + others 1.62 (4/2468) 0 (0/1539)
7–10 d MMRV 4 1.74 (2/1151) 0.99 (−3.08, 5.06) 1.84 (0.37, 9.05) 4 0 (0/1151) 0 (−3.31, 3.31) 1.44 (0.22, 9.61)
MMRV + others 0.65 (1/1539) 0 (0/1539)
MMRV + others vs. others
0–28/42 d MMRV + others 5 1.59 (3/1886) 1.70 (−1.43, 4.82) 2.52 (0.63, 10.11) 6 0.38 (1/2651) 0.37 (−1.61, 2.35) 1.46 (0.34, 6.25)
Others 0 (0/2745) 0 (0/3534)
7–10 d MMRV + others 5 0.53 (1/1886) 0.35 (−2.12, 2.81) 1.41 (0.28, 7.19) 6 0.38 (1/2651) 0.37 (−1.61, 2.35) 1.46 (0.34, 6.25)
Others 0 (0/2745) 0 (0/3534)

Second MMRV dose MMRV vs. MMR + V

0–42 d MMRV 3 0 (0/513) −2.23 (−11.02, 6.56) 0.50 (0.06, 3.84) 3 0 (0/693) −2.12 (−10.71, 6.47) 0.43 (0.06, 3.26)
MMR + V 2.38 (1/420) 2.38 (1/420)
7–10 d MMRV 3 0 (0/513) 0 (−7.64, 7.64) 0.80 (0.09, 7.42) 3 0 (0/693) 0 (−7.54, 7.54) 0.66 (0.08, 5.79)
MMR + V 0 (0/420) 0 (0/420)
MMRV vs. MMR
0–42 d MMRV 2 1.43 (1/700) −1.89 (−11.50, 7.72) 0.41 (0.05, 3.27) 5 0.20 (1/5119) 0.21 (−1.91, 2.33) 0.47 (0.09, 2.32)
MMR 3.28 (1/305) 0 (0/1575)
7–10 d MMRV 2 1.43 (1/700) 1.20 (−6.30, 8.69) 0.80 (0.07, 9.20) 5 0.20 (1/5119) 0.21 (−1.91, 2.33) 0.47 (0.09, 2.32)
MMR 0 (0/305) 0 (0/1575)
MMRV vs. MMRV + others
0–28 d MMRV 1 3.10 (1/323) 0.07 (−8.41, 8.54) 1.02 (0.06, 16.26) 1 0 (0/323) 0 (−5.98, 5.98) 1.02 (0.02, 51.33)
MMRV + others 3.03 (1/330) 0 (0/330)
7–10 d MMRV 1 0 (0/323) 0 (−5.98, 5.98) 1.02 (0.02, 51.33) 1 0 (0/323) 0 (−5.98, 5.98) 1.02 (0.02, 51.33)
MMRV + others 0 (0/1453) 0 (0/330)
MMRV + others vs. others
0–28 d MMRV + others 1 0 (0/347) 0 (−4.09, 4.09) 4.18 (0.08, 210.43) 1 0 (0/347) 0 (−4.09, 4.09) 4.18 (0.08, 210.42)
Others 0 (0/1453) 0 (0/1453)
7–10 d MMRV + others 1 0 (0/347) 0 (−4.09, 4.09) 4.18 (0.08, 210.43) 1 0 (0/347) 0 (−4.09, 4.09) 4.18 (0.08, 210.42)
Others 0 (0/1453) 0 (0/1453)

Note: No heterogeneity was found in any comparisons (I2 = 0%). 95%CI, 95% confidence interval; d, days; MMRV, measles–mumps–rubella–varicella vaccine; MMR, measles–mumps–rubella vaccine; RD, risk difference; RR, risk
ratio; V, varicella vaccine.
Comparison categories (1) MMRV vs. MMR + V, children received MMRV vaccine alone versus those received MMR and V vaccines simultaneously; (2) MMRV vs. MMR, children received MMRV vaccine versus those received
MMR vaccine; (3) MMRV vs. MMRV + others, children received MMRV vaccine alone versus those received MMRV vaccine and one or more other pediatric vaccines concomitantly; (4) MMRV + others vs. Others, children received
MMRV vaccine and one or more kinds of other pediatric vaccines concomitantly vs. those received one or more other pediatric vaccines without MMRV vaccine.

9
10

JVAC 16538 1–14

G Model
review and meta-analysis. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.06.009
Please cite this article in press as: Ma S-J, et al. Risk of febrile seizure after measles–mumps–rubella–varicella vaccine: A systematic

Table 3
Characteristics of included post-marketing observational safety studies.

Study Country Study design NOS Data MMRV Comparison Total number Time period of Age range % of male Duration of Incidence Incidence Adjustments
score source vaccine categories vaccination follow-up of seizure of febrile
type (‰) seizure (‰)

Jacobsen USA Post-licensure 9 KPSC ProQuad MMRV 31,298 2006.02–2007.06 12–60 ma 51 0–30 d ND 1.41 Age, sex and
[51] retrospective calendar date of
cohort study vaccination
MMR + V 31,298 2003.11–2006.01 ND 1.28
Klein [52] USA Retrospective 9 VSD ProQuad MMRV 83,107 2006.01–2008.10 12–23 m NR 0–42 2.27 1.93 Age, site, calendar
cohort study d/0–42 d year and
respiratory virus
season
MMR + V 376,354 2000.01–2008.10 1.59 0.62
MMR 145,302 1.04 ND
V 107,744 1.20 ND

ARTICLE IN PRESS
Klein [53] USA Retrospective 9 VSD ProQuad MMRV 86,750 2006.01–2008.10 4–6 y NR 0–42 0.22 0.01 Vaccine exposure,
cohort study d/7–10 d VSD site, age, year,
and influenza
season

S.-J. Ma et al. / Vaccine xxx (2015) xxx–xxx

MMR + V 67,438 2000.01–2008.10 0.15 0
MMR 479,311 0.21 ND
V 80,985 0.07 ND
Rowhani- USA Retrospective 8 VSD NR MMRV 120,377 2001.01–2011.12 12–23 m 51 0–42 d 2.68 ND Sex, respiratory
Rahbar cohort study virus season,
[12] calendar year and
VSD site
MMR + V 584,987 2.10 ND
MMR 134,984 1.92 ND
Gavrielov- Israel Retrospective 9 CHS Priorix- MMRV 8344 2008.09–2009.12 10–24 m 51 0–40 d ND 2.28 Age at vaccination,
Yusim cohort study Tetra ethnicity, vaccine
[54] type, preterm birth
MMR 90,294 2005.01–2008.08 ND 2.19
Hambidge USA Self-controlled NA VSD NR MMRV 323,247 2004–2010 12–23 m NR 7–10 d ND ND NA
[55] case series
MMR + V ND ND
MacDonald Canada Retrospective, 9 CIHI Priorix- MMRV 96,686 mid-2010 – 2012 12–23 m NR 0–42 d 2.95 ND Age and calendar
[56] population- Tetra year
based cohort
study
MMR + V 181,088 2006–2012 2.03 ND
Schink [57] Germany Matched 9 GePaRD Priorix- MMRV 82,656 2006.01–2008.12 0–58 mb 51 0–30 d ND 1.20 Sex, age, month of
cohort study Tetra vaccination and
statutory health
insurance
MMR + V 32,370 ND 0.93
MMR 111,241 ND 0.84

Note: d, days; m, months; y, years; NA, not applicable; ND, no data; NOS, the Newcastle-Ottawa Scale; NR, not reported; MMRV, measles–mumps–rubella–Varicella vaccine; MMR, measles–mumps–rubella vaccine; V, varicella
vaccine; Priorix-Tetra/ProQuad, MMRV vaccine.
Data source: KPSC, Kaiser Permanente Southern California; VSD, Vaccine Safety Datalink; CIHI, Canadian Institute for Health Information databases; GePaRD, German Pharmacoepidemiological Research Database; CHS, Clalit
Health Services’ data warehouse in Israel.
Age range:
a
, 99% were between the ages of 12–23 months in both groups, considered to be 12–23 months group in the subgroup analysis.
b
, ≥90% were between 11 and 23 months, considered to be 11 and 23 months group in the subgroup analysis.
G Model
JVAC 16538 1–14 ARTICLE IN PRESS
S.-J. Ma et al. / Vaccine xxx (2015) xxx–xxx 11

Comparison categories: MMRV vs. MMR + V, children received MMRV vaccine alone versus those received MMR and V vaccines simultaneously; MMRV vs. MMR, children received MMRV vaccine versus those received MMR
4. Discussion 335

2.36 (1.03, 5.38)A

2.16 (1.01, 4.64)A

2.32 (1.49, 3.60)

1.27 (0.98, 1.66)

1.04 (0.65, 1.66)
2.40 (1.40, 4.10)

1.40 (1.00, 1.90)

This systematic review synthesized evidences about the risk 336

RR (95% CI)
of febrile seizure after MMRV vaccine in 39 studies involving 337

more than 3,240,000 children. We used systematic strategy and 338

broad search terms in multiple databases and related websites to 339

identify as many published and unpublished studies as possible. 340

We also used rigorous methods to extract and appraise the data. 341

Gavrielov-Yusim, 2014♦
Gavrielov-Yusim, 2014

Gavrielov-Yusim, 2014

Note: d, days following vaccination; MMRV, measles–mumps–rubella–varicella vaccine; MMR, measles–mumps–rubella vaccine; V, varicella vaccine; RR, relative risk; 95% CI, 95% confidence interval.
Considering different comparison groups and corresponding clini- 342

cal significance, hierarchical analyses of four comparisons involving 343

MMRV vs. MMR

different follow-up periods and vaccinated ages were employed 344

Schink, 2014
Schink, 2014
directly. 345

In the risk windows of 0–42/56 days and 7–10 days after first
Studies

Pooled 346

Pooled
MMRV dose, pooled incidence of febrile seizure and vaccine related 347

febrile seizure was respectively under 2.61‰ and 0.81‰ in all 348

groups, and it was numerically higher in MMRV groups than cor- 349

responding MMR + V and MMR groups (Table 2). The incidence of

2.33 (0.10, 57.25)
350
1.46 (1.11, 1.92)A
2.04 (1.44, 2.90)A

2.20 (1.04, 4.65)A

1.10 (0.72, 1.69)A

1.37 (1.12, 1.69)

1.77 (1.08, 2.91)

1.50 (0.90, 2.30)

1.50 (0.80, 3.00)

febrile seizure was similar or lower within 0–28/42 days after sec- 351

, study was not included in the pooled analysis due to the same data source with others; A , adjusted RR; , time period was 0–30 days; ♦, time period was 0–40 days.
ond MMRV vaccine dose as compared to the first dose in most
RR (95% CI)

352

studies, which was in line with the lower fever rates after the sec- 353

ond MMRV vaccinated dose in many RCTs [26–31,33,35,49,50,59]. 354

Use of MMRV vaccine as second dose has lower risk of febrile 355
MMRV vs. MMR + V

seizure than the first dose in healthy children. 356

Jacobsen, 2009

Multiple vaccines for children are recommended in similar age

Jacobsen, 2009

357
Febrile seizure

Schink, 2014
Schink, 2014

range by public health authorities and professional medical soci-

Klein, 2012
Klein, 2010

Klein, 2010

358

eties [48]. To comply with current recommended immunization 359

Studies

Pooled

Pooled

schedules, it is necessary to administer some of these vaccines 360

simultaneously at different body sites. Our results did not find 361

that the receipt of concomitant use of MMRV vaccine with other 362

pediatric vaccines was a significant predictor of febrile seizure. Co- 363

2.19 (1.77, 2.71)
3.21 (2.20, 4.67)
1.90 (1.43, 2.53)

1.40 (1.19, 1.65)

2.46 (0.76, 7.97)

1.06 (0.65, 1.73)

administer MMRV vaccine and other routine pediatric vaccines at 364

the same time may be desirable in terms of both convenience and

RR (95% CI)

365

compliance, thus ensuring high coverage rates [34]. 366

Analyses of thirty RCTs involving total of 40,000 subjects sug- 367

gested that no significant differences in the incidence of febrile 368

seizure or vaccine related febrile seizure were found in any compar- 369
Rowhani-Rahbar, 2013

Rowhani-Rahbar, 2013

isons after any doses, in the risk windows of 0–28/42/56 days and 370

7–10 days (Table 2). The numerical differences of pooled incidence 371
MMRV vs. MMR

between groups listed in Table 2, especially those caused by one 372

reported case, might partly attribute to poor methodological qual-

Klein, 2010

Klein, 2010

373
Klein, 2012
Klein, 2012

ity of smaller studies and occasional differences of some factors, 374

Studies
Summary risk estimates and 95% CIs for post-marketing observational safety studies.

potential factors included familial tendency, day care attendance, 375

complications at birth and developmental delay, viral diseases and 376

so on [60]. 377

The pooled database of clinical trials did not show a difference.

7.00 (0.38, 129.95)

378
1.98 (1.43, 2.73)A

1.42 (1.11, 1.81)A

1.21 (0.93, 1.58)A

1.99 (1.30, 3.05)A

1.28 (1.13, 1.44)

1.27 (1.14, 1.42)

1.48 (0.69, 3.18)

2.00 (1.63, 2.45)

2.00 (1.66, 2.40)

Potential publication bias was found in the analyses of pooled RRs 379

of vaccine related febrile seizure in MMRV vs. MMR + V comparison.

RR (95% CI)

380

In spite of the sensitivity of both Begg and Egger tests is generally 381

low in meta-analysis based on less than 20 trials [61], potential pub- 382

lication bias may partly attribute to low incidence, limited sample 383

size, low possibility of reporting and different definition of vaccine 384

related febrile seizure. Moreover, data in several RCTs [27,35,38,50]

Rowhani-Rahbar, 2013

Rowhani-Rahbar, 2013

385

were not specific, we failed to acquire exact numbers of febrile

MMRV vs. MMR + V

386
MacDonald, 2014

MacDonald, 2014

seizure cases reported during 0–42 days post-vaccination period 387

following each dose (dose 1 and dose 2, respectively) in each group

Klein, 2010

Klein, 2010

388
Klein, 2012
Klein, 2012

of these RCTs, which reduced the sample size that could be com- 389
Seizure

Studies

Pooled

Pooled

bined. As serious problems are rare, it is hard to confirm or debunk 390

vaccine-associated risks [62]. In addition to post-marketing surveil- 391

lance, substantially larger and multicenter clinical trials are still 392

required to provide the highest standard of evidences to assess the 393

relative risk of febrile seizure after MMRV vaccine. Time of onset,

(vaccination age)

394
10–24 months

complication, duration, prognosis and outcome of each case should 395

be reported in detail.
4–6 years
Subgroup

396

Data of 8 post-marketing observational safety surveillance stud-

5–12 d
7–10 d

0–42 d

7–10 d
0–42 d

vaccine.

397
Table 4

ies of MMRV vaccine suggested an approximately 2-fold increase 398

in risk for seizure or febrile seizure during 7–10 days or 5–12 399

Please cite this article in press as: Ma S-J, et al. Risk of febrile seizure after measles–mumps–rubella–varicella vaccine: A systematic
review and meta-analysis. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.06.009
G Model
JVAC 16538 1–14 ARTICLE IN PRESS
12 S.-J. Ma et al. / Vaccine xxx (2015) xxx–xxx

400 days after vaccination, among children aged 10–24 months who [68]. Moreover, although all the 42 reported febrile seizure cases 466

401 received the first dose of MMRV vaccine compared with those recovered without any sequelae in our included clinical studies 467

402 received the first dose of MMR vaccine administered with or with- involved to prognosis [27–37,39,40,45,48,49], it was unable to pre- 468

403 out varicella vaccine [12,51,52,54,56,57], the approximately 2-fold dict authentic and comprehensive prognosis of febrile seizures after 469

404 increased risk resulted in an estimated one extra febrile seizure per MMRV vaccine. Studies are still needed to identify the real risk and 470

405 2300–2600 MMRV vaccine doses [51,52]. Compared with no vac- prognosis of febrile seizures after MMRV vaccine in order to guide 471

406 cination, MMR vaccine was associated with one additional febrile its application. 472

407 seizure for every 3000–4000 children aged under 7 years in a risk Vaccines are biological products with biological effects. We can 473

408 window of 8–14 days after MMR vaccination [63]. Furthermore, never make them 100% safe [62]. It is important to emphasize 474

409 the seizure risk during 0–42 days after vaccination was also signif- that the benefits of vaccination still greatly outweigh its risks [62]. 475

410 icantly higher after MMRV vaccine, compared with MMR + V and The recommended immunization schedule for children needs to be 476

411 MMR with lower RRs (RR 1.27; 95% CI 1.14, 1.42 and RR 1.40; 95% formulated using the best available data to maximize benefit and 477

412 CI 1.19, 1.65, respectively). However, the significant increase of minimize risk [69]. We highlight that, providers who are consider- 478

413 febrile seizure in children aged 10–24 months during 0–30/40/42 ing administering MMRV vaccine to children aged 12–47 months 479

414 days after MMRV vaccine was only found when it was compared should take extra cautions and provide parents or caregivers con- 480

415 with MMR + V vaccines (RR 1.37; 95% CI 1.12, 1.69). Considering the sultation on both benefits and risks. Decisions should be made by 481

416 increased overall risk for seizure or febrile seizure after MMRV vac- them on a case-by-case basis [13,26]. 482

417 cination may be caused by the excess of seizures or febrile seizures

418 that occur during the second week after immunization, MMRV vac- 5. Conclusion 483
419 cine may cause a discontinuous increased risk for seizure or febrile
420 seizure [52]. Clinical studies did not find any difference in the incidence of 484
421 The significantly higher risks among children aged 10–24 febrile seizure or vaccine related febrile seizure between MMRV 485
422 months suggested that parents of first MMRV vaccine recipients vaccine recipients and MMR + V, MMR, MMRV + others vaccine 486
423 should be counseled about the possibility of seizure or febrile recipients after any doses, in the risk windows of 0–28/42/56 487
424 seizure within 0–42 days after vaccination (especially within the days and 7–10 days. However, an approximately 2-fold increase 488
425 second week), and educated on timing. The cases usually need in risk for seizure or febrile seizure during 7–10 days or 5–12 days 489
426 to seek medical attention, which commonly includes a visit to after MMRV vaccination among children aged 10–24 months was 490
427 an emergency department. Consistent with the biologic window demonstrated in several post-marketing surveillance studies. More 491
428 of vulnerability for febrile seizure in children (approximately 97% post-marketing restudies based on rigorously prospective study 492
429 occur in children aged ≤4 years) [64], data from a post-marketing design are needed to confirm the findings. 493
430 study suggested no evidence that MMRV vaccine was associated
431 with an elevated risk of febrile seizure among children aged 4–6 Conflicts of interest 494
432 years old during 7–10 days or 0–42 days after vaccination [53].
433 Use of MMRV vaccine at age ≥4 years still has been recommended None. 495
434 as a preference over separate injections of its equivalent compo-
435 nent vaccines [13]. As approximately one third of children with a
Funding 496
436 first febrile seizure will have recurrent febrile seizures [65], chil-
437 dren with a known personal or family history of febrile seizure
The authors have not accepted any financial support for this 497
438 should receive the MMR + V vaccines instead of MMRV vaccine for
work. 498
439 any doses.
440 Significantly higher risks which were found in post-marketing
Author contributions 499
441 observational studies, rather than clinical trials, might partly due
442 to limited population and condition, protocol-specified vaccinated
All authors: analysis and interpretation of data, and preparation 500
443 ages and intervals studied in RCTs. Following licensure and once in
of manuscript. Qing Chen and Shu-Juan Ma: Conception and design. 501
444 routine use, MMRV vaccine can be used in millions of people, SAEs
445 like febrile seizures that occur too infrequently to detect in pre-
446 licensure RCTs may be identified and further studied. In addition, References 502

447 when rigorous criteria like that only healthy children are included
[1] Stuck B, Stehr K, Bock HL. Concomitant administration of varicella vaccine with 503
448 in RCTs, more febrile seizures may be reported among general chil- combined measles, mumps, and rubella vaccine in healthy children aged 12 to 504
449 dren in observational studies. Just as Müeller at al. emphasized that 24 months of age. Asian Pac J Allergy Immunol 2002;20:113–20. 505
[2] Marin M, Guris D, Chaves SS, Schmid S, Seward JF. Prevention of varicella: rec- 506
450 non-randomized studies, which can generally be larger, may be in
ommendations of the Advisory Committee on Immunization Practices (ACIP). 507
451 a better position to detect infrequent complications [66]. However, MMWR Recomm Rep 2007;56:1–40. 508
452 bias is more common in retrospective studies than in prospec- [3] Kreth HW, Lee BW, Kosuwon P, Salazar J, Gloriani-Barzaga N, Bock HL, et al. 509

453 tive studies. More post-marketing restudies based on rigorously Sixteen years of global experience with the first refrigerator-stable varicella 510
vaccine (Varilrix). BioDrugs 2008;22:387–402. 511
454 prospective study design are needed to confirm the findings, like [4] CDC. Licensure of a combines live attenuated measles, mumps, rubella, and 512
455 real word researches, which can provide a more complete picture varicella vaccine. Morb Mortal Wkly Rep 2005;57(47):1212–4. 513

456 of the real world clinical outcomes. [5] Knuf M, Faber J, Barth I, Habermehl P. A combination vaccine against measles, 514
mumps, rubella and varicella. Drugs Today (Barc) 2008;44:279–92. 515
457 Vaccine-induced febrile seizure is a rare phenomenon that [6] Kuter BJ, Brown ML, Hartzel J, Williams WR, EvesiKaren A, Black S, et al. Safety 516
458 occurs mainly in genetically predisposed children [16]. The inci- and immunogenicity of a combination measles, mumps, rubella and varicella 517
459 dence of seizure or febrile seizure reported in observational studies vaccine (ProQuad). Hum Vaccin 2006;2:205–14. 518
[7] Dhillon S, Curran MP. Live attenuated measles, mumps, rubella, and varicella 519
460 during the second week and 0–42 days after MMRV immunization zoster virus vaccine (Priorix-Tetra). Paediatr Drugs 2008;10:337–47. 520
461 was respectively about 1.00‰ and 2.20‰ in children aged 10–24 [8] Hanna C, Volker S, Fred Z, Susanna E, Martine D, Paul W. A combined measles, 521
462 months. Although there is no evidence suggesting that occurrence mumps, rubella and varicella vaccine (Priorix-Tetra (TM)): immunogenicity 522
and safety profile. Vaccine 2009;27:6504–11. 523
463 of febrile seizures increase the risk of death, brain damage, or lean-
[9] Klopfer SO, Stek JE, Petrecz M, Reisinger KS, Black SB, Goveia MG, et al. Analysis 524
464 ing disorders [10,63,67], parents generally consider febrile seizure of safety data in children after receiving two doses of ProQuad((R)) (MMRV). 525
465 to be a more severe adverse event than what physicians believe Vaccine 2014;32:7154–60. 526

Please cite this article in press as: Ma S-J, et al. Risk of febrile seizure after measles–mumps–rubella–varicella vaccine: A systematic
review and meta-analysis. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.06.009
G Model
JVAC 16538 1–14 ARTICLE IN PRESS
S.-J. Ma et al. / Vaccine xxx (2015) xxx–xxx 13

527 [10] Lorin MI. Fever: pathogenesis and treatment. In: Feigin RD, Cherry JD, editors. [38] Vesikari T, Karvonen A, Lindblad N, Korhonen T, Lommel P, Willems P, et al. 613
528 Textbook of pediatric infectious diseases, vol. 1, 4th ed. Philadelphia, PA: WB Safety and immunogenicity of a booster dose of the 10-valent pneumococcal 614
529 Saunders Company; 1998. p. 89–95. nontypeable Haemophilus influenzae protein D conjugate vaccine coadminis- 615
530 [11] Jankowiak J, Malow B. Seizures in children with fever: generally good outcome. tered with measles–mumps–rubella–varicella vaccine in children aged 12 to 616
531 Neurology 2003;60:E1–2. 16 months. Pediatr Infect Dis J 2010:e47–56. 617
532 [12] Rowhani-Rahbar A, Fireman B, Lewis E, Nordin J, Naleway A, Jacobsen SJ, et al. [39] Leonardi M, Bromberg K, Baxter R, Gardner JL, Klopfer S, Nicholson O, et al. 618
533 Effect of age on the risk of fever and seizures following immunization with Immunogenicity and safety of MMRV and PCV-7 administered concomitantly 619
534 measles-containing vaccines in children. JAMA Pediatr 2013;167:1111–7. in healthy children. Pediatrics 2011:e1387–94. 620
535 [13] Marin M, Broder KR, Temte JL, Snider DE, Seward JF. Use of combination [40] Rümke HC, Loch HP, Hoppenbrouwers K, Vandermeulen C, Malfroot A, Helm K, 621
536 measles, mumps, rubella, and varicella vaccine: recommendations of the et al. Immunogenicity and safety of a measles–mumps–rubella–varicella vac- 622
537 Advisory Committee on Immunization Practices (ACIP). MMWR Recomm Rep cine following a 4-week or a 12-month interval between two doses. Vaccine 623
538 2010;59:1–12. 2011:3842–9. 624
539 [14] Friedman MJ, Sharieff GQ. Seizures in children. Pediatr Clin North Am [41] Vesikari T, Karvonen A, Bianco V, Van der Wielen M, Miller J. Tetrava- 625
540 2006;53:257–77. lent meningococcal serogroups A, C, W-135 and Y conjugate vac- 626
541 [15] von Spiczak S, Helbig I, Drechsel-Baeuerle U, Muhle H, van Baalen A, van Kem- cine is well tolerated and immunogenic when co-administered with 627
542 pen MJ, et al. A retrospective population-based study on seizures related to measles–mumps–rubella–varicella vaccine during the second year of life: an 628
543 childhood vaccination. Epilepsia 2011;52:1506–12. open, randomized controlled trial. Vaccine 2011;29:4274–84. 629
544 [16] Principi N, Esposito S. Vaccines and febrile seizures. Expert Rev Vaccin [42] Blatter MM, Klein NP, Shepard JS, Leonardi M, Shapiro S, Schear M, et al. 630
545 2013;12:885–92. Immunogenicity and safety of two tetravalent (measles, mumps, rubella, vari- 631
546 [17] Jadad AR, Moore RA, Carroll D, Jenkinson C, Reynolds DJ, Gavaghan DJ, et al. cella) vaccines coadministered with hepatitis a and pneumococcal conjugate 632
547 Assessing the quality of reports of randomized clinical trials: is blinding nec- vaccines to children twelve to fourteen months of age. Pediatr Infect Dis J 633
548 essary. Control Clin Trials 1996;17:1–12. 2012;31:e133–40. 634
549 [18] Stang A. Critical evaluation of the Newcastle-Ottawa scale for the assessment [43] Klein NP, Shepard J, Bedell L, Odrljin T, Dull P. Immunogenicity and safety of 635
550 of the quality of nonrandomized studies in meta-analyses. Eur J Epidemiol a quadrivalent meningococcal conjugate vaccine administered concomitantly 636
551 2010;25:603–5. with measles, mumps, rubella, varicella vaccine in healthy toddlers. Vaccine 637
552 [19] Wells GA, O’Connell SB, Peterson D, Welch J, Losos V, Tugwell MP. The 2012:3929–36. 638
553 Newcastle–Ottawa Scale (NOS) for assessing the quality of nonrandomised [44] Ruger G, Gabutti G, Rumke H, Rombo L, Bernaola E, Diez-Domingo J, et al. Safety 639
554 studies in meta-analyses; 2014. Available: http://wwwohrica/programs/ of a 2-dose regimen of a combined measles, mumps, rubella and varicella live 640
555 clinical epidemiology/oxfordasp [accessed 05.11.14]. vaccine manufactured with recombinant human albumin. Pediatr Infect Dis J 641
556 [20] Song F, Sheldon TA, Sutton AJ, Abrams KR, Jones DR. Methods for exploring 2012;31:1166–72. 642
557 heterogeneity in meta-analysis. Eval Health Prof 2001;24:126–51. [45] Vesikari L, Becker T, Gajdos V, Fiquet A, Thomas S, Richard P, et al. Immuno- 643
558 [21] DerSimonian R, Laird N. Meta-analysis in clinical trials. Control Clin Trials genicity and safety of a two-dose regimen of a combined measles, mumps, 644
559 1986;7:177–88. rubella and varicella live vaccine (ProQuad (R)) in infants from 9 months of 645
560 [22] Mantel N, Haenszel W. Statistical aspects of the analysis of data from retro- age. Vaccine 2012;30:3082–9. 646
561 spective studies of disease. J Natl Cancer Inst 1959;22:719–48. [46] Huang LM, Lee BW, Chan PC, Povey M, Henry O. Immunogenicity and safety of 647
562 [23] Higgins JP, Thompson SG. Quantifying heterogeneity in a meta-analysis. Stat combined measles–mumps–rubella–varicella vaccine using new measles and 648
563 Med 2002;21:1539–58. rubella working seeds in healthy children in Taiwan and Singapore: a phase II, 649
564 [24] Egger M, Davey Smith G, Schneider M, Minder C. Bias in meta-analysis detected randomized, double-blind trial. Hum Vaccin Immunother 2013;9:1308–15. 650
565 by a simple, graphical test. BMJ 1997;315:629–34. [47] Vesikari T, Esposito S, Prymula R, Ypma E, Kohl I, Toneatto D, et al. Immuno- 651
566 [25] Begg CB, Mazumdar M. Operating characteristics of a rank correlation test for genicity and safety of an investigational multicomponent, recombinant, 652
567 publication bias. Biometrics 1994;50:1088–101. meningococcal serogroup B vaccine (4CMenB) administered concomitantly 653
568 [26] GlaxoSmithKline. GlaxoSmithKline clinical trial register: 208136 (MeMuRu- with routine infant and child vaccinations: results of two randomised trials. 654
569 OKA). Available: http://wwwgsk-clinicalstudyregistercom/ (accessed Lancet 2013;381:825–35. 655
570 16.10.14). [48] Yetman RJ, Shepard JS, Duke A, Stek JE, Petrecz M, Klopfer SO, et al. Concomitant 656
571 [27] Health NIo ClinicalTrials.gov, available: http://clinicaltrialsgov/ (accessed administration of hepatitis A vaccine with measles/mumps/rubella/varicella 657
572 12.10.14). and pneumococcal vaccines in healthy 12- to 23-month-old children. Hum 658
573 [28] Shinefield H, Black S, Digilio L, Reisinger K, Blatter M, Gress JO, et al. Evaluation Vaccin Immunother 2013;9:1691–7. 659
574 of a quadrivalent measles, mumps, rubella and varicella vaccine in healthy [49] Cha SH, Shin SH, Lee TJ, Kim CH, Povey M, Kim HM, et al. Immunogenicity and 660
575 children. Pediatr Infect Dis J 2005;24:665–9. safety of a tetravalent measles–mumps–rubella–varicella vaccine: an open- 661
576 [29] Shinefield H, Black S, Williams WR, Marchant C, Reisinger K, Stewart T, et al. labeled, randomized trial in healthy Korean children. Clin Exp Vaccine Res 662
577 Dose–response study of a quadrivalent measles, mumps, rubella and varicella 2014;3:91–9. 663
578 vaccine in healthy children. Pediatr Infect Dis J 2005;24:670–5. [50] Prymula R, Bergsaker MR, Esposito S, Gothefors L, Man S, Snegova 664
579 [30] Knuf M, Habermehl P, Zepp F, Mannhardt W, Kuttnig M, Mutto- N, et al. Protection against varicella with two doses of combined 665
580 nen P, et al. Immunogenicity and safety of two doses of tetravalent measles–mumps–rubella–varicella vaccine versus one dose of monovalent 666
581 measles–mumps–rubella–varicella vaccine in healthy children. Pediatr Infect varicella vaccine: a multicentre, observer-blind, randomised, controlled trial. 667
582 Dis J 2006:12–8. Lancet 2014;383:1313–24. 668
583 [31] Lieberman JM, Williams WR, Miller JM, Black S, Shinefield H, Henderson F, et al. [51] Jacobsen SJ, Ackerson BK, Sy LS, Tran TN, Jones TL, Yao JF, et al. Observational 669
584 The safety and immunogenicity of a quadrivalent measles, mumps, rubella and safety study of febrile convulsion following first dose MMRV vaccination in a 670
585 varicella vaccine in healthy children – a study of manufacturing consistency managed care setting. Vaccine 2009;27:4656–61. 671
586 and persistence of antibody. Pediatr Infect Dis J 2006;25:615–22. [52] Klein NP, Fireman B, Yih WK, Lewis E, Kulldorff M, Ray P, et al. 672
587 [32] Shinefield H, Black S, Thear M, Coury D, Reisinger K, Rothstein E, et al. Safety Measles–mumps–rubella–varicella combination vaccine and the risk of febrile 673
588 and immunogenicity of a measles, mumps, rubella and varicella vaccine given seizures. Pediatrics 2010;126:e1–8. 674
589 with combined Haemophilus influenzae type b conjugate/hepatitis B vaccines [53] Klein NP, Lewis E, Baxter R, Weintraub E, Glanz J, Naleway A, et al. Measles- 675
590 and combined diphtheria-tetanus-acellular pertussis vaccines. Pediatr Infect containing vaccines and febrile seizures in children age 4 to 6 years. Pediatrics 676
591 Dis J 2006;25:287–92. 2012;129:809–14. 677
592 [33] Goh P, Lim FS, Han HH, Willems P. Safety and immunogenicity of early [54] Gavrielov-Yusim N, Hoshen M, Singer SR, Neumann L, Balicer RD. The weight 678
593 vaccination with two doses of tetravalent measles–mumps–rubella–varicella of MMRV-related febrile convulsions among other clinical factors contributing 679
594 (MMRV) vaccine in healthy children from 9 months of age. Infection 2007: to febrile convulsions in children. Vaccine 2014;32:4954–9. 680
595 326–33. [55] Hambidge SJ, Newcomer SR, Narwaney KJ, Glanz JM, Daley MF, Xu S, et al. 681
596 [34] Zepp F, Behre U, Kindler K, Laakmann KH, Pankow-Culot H, Mannhardt- Timely versus delayed early childhood vaccination and seizures. Pediatrics 682
597 Laakmann W, et al. Immunogenicity and safety of a tetravalent 2014;133:e1492–9. 683
598 measles–mumps–rubella–varicella vaccine co-administered with a booster [56] MacDonald SE, Dover DC, Simmonds KA, Svenson LW. Risk of febrile seizures 684
599 dose of a combined diphtheria-tetanus-acellular pertussis-hepatitis B- after first dose of measles–mumps–rubella–varicella vaccine: a population- 685
600 inactivated poliovirus-Haemophilus influenzae type b conjugate vaccine in based cohort study. CMAJ 2014;186:824–9. 686
601 healthy children aged 12–23 months. Eur J Pediatr 2007:857–64. [57] Schink T, Holstiege J, Kowalzik F, Zepp F, Garbe E. Risk of febrile convulsions 687
602 [35] Schuster V, Otto W, Maurer L, Tcherepnine P, Pfletschinger U, Kindler K, after MMRV vaccination in comparison to MMR or MMR+V vaccination. Vaccine 688
603 et al. Immunogenicity and safety assessments after one and two doses of a 2014;32:645–50. 689
604 refrigerator-stable tetravalent measles–mumps–rubella–varicella vaccine in [58] Bonhoeffer J, Menkes J, Gold MS, de Souza-Brito G, Fisher MC, Halsey N, et al. 690
605 healthy children during the second year of life. Pediatr Infect Dis J 2008:724–30. Generalized convulsive seizure as an adverse event following immunization: 691
606 [36] Halperin SA, Ferrera G, Scheifele D, Predy G, Stella G, Cuccia M, et al. Safety case definition and guidelines for data collection, analysis, and presentation. 692
607 and immunogenicity of a measles–mumps–rubella–varicella vaccine given as Vaccine 2004;22:557–62. 693
608 a second dose in children up to six years of age. Vaccine 2009:2701–6. [59] Nolan T, McIntyre P, Roberton D, Descamps D. Reactogenicity and immuno- 694
609 [37] Knuf M, Zepp F, Meyer CU, Habermehl P, Maurer L, Burow HM, et al. Safety, genicity of a live attenuated tetravalent measles–mumps–rubella–varicella 695
610 immunogenicity and immediate pain of intramuscular versus subcutaneous (MMRV) vaccine. Vaccine 2002:281–9. 696
611 administration of a measles–mumps–rubella–varicella vaccine to children [60] Bethune P, Gordon K, Dooley J, Camfield C, Camfield P. Which child will have a 697
612 aged 11–21 months. Eur J Pediatr 2010:925–33. febrile seizure. Am J Dis Child 1993;147:35–9. 698

Please cite this article in press as: Ma S-J, et al. Risk of febrile seizure after measles–mumps–rubella–varicella vaccine: A systematic
review and meta-analysis. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.06.009
G Model
JVAC 16538 1–14 ARTICLE IN PRESS
14 S.-J. Ma et al. / Vaccine xxx (2015) xxx–xxx

699 [61] Sterne JA, Egger M, Smith GD. Systematic reviews in health care: Investi- [66] Knottnerus JA, Tugwell P. Real world research. J Clin Epidemiol 710
700 gating and dealing with publication and other biases in meta-analysis. BMJ 2010;63:1051–2. 711
701 2001;323:101–5. [67] Practice parameter: long-term treatment of the child with simple febrile 712
702 [62] Kwok R. Vaccines, the real issues in vaccine safety. Nature 2011;473:436–8. seizures. American Academy of Pediatrics. Committee on Quality Improvement, 713
703 [63] Barlow WE, Davis RL, Glasser JW, Rhodes PH, Thompson RS, Mullooly JP, et al. Subcommittee on Febrile Seizures. Pediatrics 1999;103:1307–9. 714
704 The risk of seizures after receipt of whole-cell pertussis or measles, mumps, [68] O’Leary ST, Suh CA, Marin M. Febrile seizures and measles–mumps– 715
705 and rubella vaccine. N Engl J Med 2001;345:656–61. rubella–varicella (MMRV) vaccine: what do primary care physicians think. 716
706 [64] Sillanpaa M, Camfield P, Camfield C, Haataja L, Aromaa M, Helenius H, et al. Vaccine 2012;30:6731–3. 717
707 Incidence of febrile seizures in Finland: prospective population-based study. [69] Endres J, Graber MA, Dachs R. Evidence to guide measles immunization recom- 718
708 Pediatr Neurol 2008;38:391–4. mendations. Am Fam Physician 2014;89:786–7. 719
709 [65] Baulac S, Gourfinkel-An I, Nabbout R, Huberfeld G, Serratosa J, Leguern E, et al.
Fever, genes, and epilepsy. Lancet Neurol 2004;3:421–30.

Please cite this article in press as: Ma S-J, et al. Risk of febrile seizure after measles–mumps–rubella–varicella vaccine: A systematic
review and meta-analysis. Vaccine (2015), http://dx.doi.org/10.1016/j.vaccine.2015.06.009