See

discussions, stats, and author profiles for this publication at: http://www.researchgate.net/publication/8345208

Pregnancy outcome following rubella

vaccination: A prospective controlled study
ARTICLE in AMERICAN JOURNAL OF MEDICAL GENETICS PART A SEPTEMBER 2004
Impact Factor: 2.05 DOI: 10.1002/ajmg.a.30225 Source: PubMed

CITATIONS

33

6 AUTHORS, INCLUDING:
Benjamin Bar-Oz

Myla Moretti

Hadassah Medical Center

SickKids

99 PUBLICATIONS 1,807 CITATIONS

84 PUBLICATIONS 1,761 CITATIONS

SEE PROFILE

SEE PROFILE

Gideon Koren
SickKids
1,609 PUBLICATIONS 32,809 CITATIONS
SEE PROFILE

Available from: Benjamin Bar-Oz

Retrieved on: 24 August 2015

American Journal of Medical Genetics 130A:52 54 (2004)

Pregnancy Outcome Following Rubella Vaccination:

A Prospective Controlled Study
Benjamin Bar-Oz,1,2 Zina Levichek,1 Myla E. Moretti,1 Corinna Mah,1 Stella Andreou,1 and Gideon Koren1*
1

The Motherisk Program, Division of Clinical Pharmacology, The Hospital for Sick Children and The University of Toronto, Canada
Department of Neonatology, Hadassah Medical Center, Jerusalem, Israel

The rubella virus is a potent human teratogen.

Because the rubella vaccine is prepared with live
virus, a high level of anxiety surrounds exposure
in pregnancy. There is relatively scarce data on
fetal risk following vaccination in pregnancy, and
all of the available data were collected retrospectively. Our objective was to examine whether
periconceptional exposure to rubella vaccine can
cause the congenital rubella syndrome, and to
compare the rate of major malformations and
developmental milestones among offspring of
women who received rubella vaccine 3 months
pre- or post-conception to an unexposed comparison group. We collected prospectively and
followed up 94 women who received rubella
vaccination 3 months pre- or post-conception
and a comparison group that consisted of 94
women who were counseled during pregnancy in
a similar manner but were not exposed to known
teratogens. The controls were matched for age,
smoking, alcohol, and drug use. Not any of the
women exposed to the vaccine gave birth to a child
with congenital rubella syndrome. Rates of major
malformations were similar in both groups as
were birth weights and developmental milestones. In contrast, the rate of therapeutic abortions was higher in the exposed group (7.4% vs. 0%)
(P < 0.05), due to fears of teratogenicity. We conclude that rubella vaccination in pregnancy
does not appear to affect pregnancy outcome in
general or cause congenital rubella syndrome in
particular. 2004 Wiley-Liss, Inc.
KEY WORDS: congenital rubella syndrome; vaccination; pregnancy outcome
INTRODUCTION
The rubella vaccine, containing a live attenuated virus, has
been licensed for general use since the late sixties. The vaccine
can be given alone or more commonly in combination with
measles and mumps vaccines (MMR). Since rubella infection
causes the congenital rubella syndrome, immunization of
postpubertal females who do not have documented evidence
of rubella immunity is essential.

*Correspondence to: Dr. Gideon Koren, M.D., FACMT, FRCPC,

Division of Clinical Pharmacology/Toxicology, The Hospital for
Sick Children, 555 University Avenue, Toronto, Ontario M5G
1X8, Canada. E-mail: gkoren@sickkids.ca
Received 12 May 2003; Accepted 22 March 2004
DOI 10.1002/ajmg.a.30225

2004 Wiley-Liss, Inc.

Presently, there is no requirement for pregnancy testing

prior to vaccination with rubella vaccine, and contraception is
not strongly advised in that context. Because half of all
pregnancies are unplanned, women may be unknowingly
pregnant at the time of vaccination, or may become pregnant
shortly thereafter. Due to the potential fetal hazards of the live
virus, the label states explicitly that the vaccine is contraindicated in pregnancy or within 3 months prior to conception.
Unfortunately, this strong labeling is interpreted by physicians and patients alike as proven evidence of teratogenicity.
The attenuated rubella-vaccine virus has been shown to
cross the placenta and was found in products of conception
as well as in the uterine cervix [Phillips et al., 1970; Vaheri
et al., 1972; Bolognese et al., 1973; Ebbin et al., 1973; Wyll and
Herrmann, 1973; Fleet et al., 1974]. Serological evidence of
congenital rubella infection was found in clinically normal infants born to mothers vaccinated in early pregnancy [Hayden
et al., 1980]. Because the attenuated virus can be found in pharyngeal excretions for at least 1 month after vaccination
[Canadian Pharmaceutical Association, 2003], it is conceivable
that it may reach the fetus even if vaccination occurs before
conception.
There are few reports assessing pregnancy outcome after
inadvertent rubella vaccination in pregnancy [Ebbin et al.,
1973; Enders, 1985; CDC, 1989; Tookey et al., 1991]. Beginning
in 1971, the Centers for Disease Control (CDC) have maintained a registry of women vaccinated during pregnancy with
different strains of rubella vaccine. From 272 susceptible
women vaccinated just before or during pregnancy, 254 (93%)
outcomes were known. No congenital rubella-like defects were
detected in any of the infants or subsequently during their
follow-up [Bart et al., 1985; Preblud and Williams, 1985; CDC,
1989; Burgess, 1990]. Similar results were reported from
Germany and the UK [Enders, 1985; Tookey et al., 1991].
However, all reports available today are retrospective, and
may be affected by selection, reporting, and recall biases. While
pharyngeal excretion of the rubella virus is known to occur
approximately 728 days after vaccination, with peak excretion around the 14th day, there is no evidence of transmission
of this excreted vaccine virus to susceptible contacts [Canadian
Pharmaceutical Association, 2003].
At the present time inadvertent rubella immunization just
before or during early pregnancy is not an accepted indication
for interruption of pregnancy [Enders, 1985; CDC, 1989].
However, it has been our clinical impression that women often
opt to have therapeutic abortions when conceiving soon after
rubella vaccination. The primary objective of this first
prospective-controlled study was to ascertain in a prospective
manner whether rubella vaccine given periconceptionally
increases the risk for congenital rubella syndrome. The secondary objectives were evaluation of rates of major malformations, rates of spontaneous and therapeutic abortions, mean
birth weight, gestational age at birth, and developmental
milestones. We were also interested in elucidating how many
women terminated an otherwise wanted pregnancy due to
fears of teratogenicity.

Pregnancy Outcome Following Rubella Vaccination

53

TABLE I. Pregnancy Outcome

Pregnancy outcome (no.)

Spontaneous abortion
Therapeutic abortion
Gestational age (no.)
Term
Preterm
Neonatal death (no.)
Birth weight
(kg, mean
SD)
Rate of malformations

Vaccinated
group

Control group

P-value

Relative risk
(95% confidence
interval)

6/94 (6.4%)
7/94 (7.4%)

8/95 (8.4%)
0/95 (0%)

0.59
0.007

0.74 (0.252.3)
2.09 (1.82.43)

74/81 (91.4 %)
7/81 (8.6%)
1/81 (1.2%)
3.34
0.64

79/87 (90.8%)
8/87 (9.2%)
0/87 (0%)
3.46
0.57

0.89
0.89
0.49
0.45

3/81 (3.7%)

3/87 (3.4%)

0.74

0.93 (0.322.7)

1.08 (0.215.5)

1/2

CI, RR.exp {
1.96[1  Ai/n1)/Ai (Ci/no:/Ci] }.

A, no. of malformed exposed; B, no. of non-malformed exposed; C, no. of malformed unexposed; D, no. of nonmalformed unexposed.

STUDY DESIGN
The Motherisk Program is a counseling service for pregnant
and lactating women, or those who plan pregnancy, and their
health professionals, on the safety/risk of drugs, chemicals,
radiation, and infections. For the purpose of the study, we
followed-up all women who had called our service requesting
information about the safety of rubella immunization in
pregnancy or 3 months prior to conception. During the initial
counseling session, details of the exposure were obtained,
including the date of rubella/MMR vaccination, reasons for
vaccination as well as maternal demographics and obstetrical
history. The type of rubella vaccine was RA 27/3 from three
different companies: MoRu-Viraten BernaTM, Berna products;
M-M-R11, Merck Frosst; Priorix, and Smithkline Beecham.
At follow-up, at least 6 months after the expected date of
birth, women were questioned regarding the course of their
pregnancy, any health issues in their children, and any
medications or chemical exposures occurring during pregnancy after the original counseling. Maternal reports were
corroborated by a report from the childs primary care
physician.
The protocol was approved by our hospital Research Ethics
Board. We received an oral consent from each participant after
the study was fully explained over the telephone.
The primary endpoint of interest was the incidence of
congenital rubella syndrome (CRS).
CRS was defined as at least one of the pathognomonic
symptoms: sensorineural hearing loss, panencephalitis, mental retardation, cardiovascular malformations (patent ductus
arteriosus, peripheral or central pulmonary stenosis, aortic or
renal arterial stenosis), and eye malformations (cataract,
glaucoma, microphthalmia) [South and Sever, 1985].
Secondary outcomes included rates of major malformations,
defined as the presence of an anomaly that has an adverse
effect on either the function or the social acceptability of the
child [Marden et al., 1964], assessment of developmental
milestones by the Denver test, rates of spontaneous or
therapeutic abortions, livebirths and stillbirths, as well as
gestational age at birth, and birthweight. The Denver developmental tool is a screening instrument which can provide
evidence of mental retardation, but not of mild delays. Hearing
was investigated by direct inquiry of the mother and evidence
from the physician caring for the child.
Because CRS is specific to fetal exposure to the virus, we
considered even a single case of CRS in the vaccinated group to
be clinically significant.
A comparison group of women exposed to non-teratogenic
drugs, counseled at similar gestational ages and followed-up
in a similar way, was also recruited and matched for age

(
2 years), smoking status, alcohol, and street drugs use. The
non-teratogenic drugs included acetaminophen, salicylate,
dental X-ray, penicillin, antihistamines. Outcomes of interest
were compared between the study and comparison group with
the t-test or MannWhitney test whenever appropriate. Rates
were compared by the Fishers exact test.
RESULTS
A total of 94 exposed women and 94 controls were followed up
prospectively. There were no statistically significant differences in maternal age, rates of smoking, alcohol consumption
and recreational drug use between the exposed and comparison group. In addition, there were no statistically significant
differences in obstetric history, gravidity, and parity.
Fifty-six (59.5%) women in the exposed group received
rubella or MMR vaccination prior to conception and 38 (40.4%)
women received it in the first trimester. Sixty-two (65.9%)
women received MMR vaccine and 28 (29.8%) received rubella
vaccine. There was no information about the type of vaccination in four women (4.3%).
The reasons for vaccination were: no titer (32; 34%), low titer
(21; 22.3%), exposure to disease (1; 1%), and 29 (31.1%) women
reported other reasons for vaccination: work requirements
(n 10), doctors recommendation (n 5), college/school requirements (n 4), no vaccination in the past (n 3), pregnancy planning (n 3), travel (n 2), military service (n 1),
immigration (n 1). Eleven women did not provide a reason for
their immunization.
There were 81 live births, 7 therapeutic abortions, and
6 miscarriages in the exposed group. The rates of therapeutic
abortions were statistically higher in the exposure group
(P 0.007) (Table I). Five out of seven elective abortions were
reported by the women to be prompted by fears of teratogenicity secondary to the vaccination. There was no difference
between two groups in mean gestational age at birth, birth
weight, or in rates of prematurity (Table I).

TABLE II. List of Major Malformations

Vaccinated group

Control group

1. Atrial septal defect

2. Excess tongue membrane that needed to
be repaired
3. Potter syndrome
1. Congenital hearing loss
2. Tongue tie, toe abnormality
3. Right inguinal hernia, webbed toes,
strabismus

54

Bar-Oz et al.
TABLE III. Developmental Milestones
Exposed (n 81)

Control (n 87)

Months
(mean
SD)

Months
(mean
SD)

Normal range
(months)

P-value

1.89
1.05
2.62
1.74
5.84
1.56
9.21
2.69
8.08
3.52
11.94
3.07

2.07
1.04
2.26
1.24
5.76
1.24
8.75
1.76
8.81
2.79
11.27
1.59

2
3
68
810
812
1215

0.33
0.32
0.77
0.58
0.98
0.55

Smile
Lift head
Sit unaided
Stand unaided
Speak first word
Walk unaided

None of the children exhibited signs of congenital rubella

syndrome. There were no differences between the two groups
in the rates of major malformations (Table I). For all three
malformed children in the exposed group, mothers received the
vaccine during the first trimester. There was one case of Potter
syndrome in the exposed group. The child died 4 hr after birth
(Table II). There were no differences in the mean age of
attainment of developmental milestones (Table III).
DISCUSSION
Although previous reports did not show increased teratogenic risk of the rubella vaccine, they were based on retrospective cohorts and registries. This is the first prospective,
controlled study of rubella vaccination in pregnancy. The
primary end point of this study was the occurrence of
congenital rubella syndrome, which could not be shown in
any of the cases exposed to the vaccine of critical importance;
the only difference between the exposed and comparison
groups was found in the rates of therapeutic abortions (seven
in the exposed group versus none in the control group).
The majority of women who chose to abort reported that the
decision was caused by the fears of teratogenicity of the
inadvertent immunization. These women chose to terminate
despite the reassuring counseling given to these women by
Motherisk. Specifically, the counseling underscores that there
is no evidence for increased teratogenic risk. It is likely that
counseling prevented many more abortions among the many
concerned mothers [Koren and Pastuszak, 1990]. This reflects
the high level of misinformation and misperception of
teratogenic risk when an exposure, believed to be teratogenic
is not having such risk. We have previously shown that such
misinformation and misperception led many women to consider termination or even to terminate otherwise wanted pregnancies [Koren and Pastuszak, 1990; Koren, 1997].
There were no statistically significant differences in neonatal outcome, birth weight and developmental milestones. Our
special interest was on speech development, since sensorineural deafness is one of the features of congenital rubella
syndrome. Age of attainment of first word was not different
between the groups.
This study has a limited power to show increased risk of
major malformations in general; however, it was powered
sufficiently for the primary endpoint of CRS, and there were no
congenital malformations consistent with congenital rubella
syndrome in the exposed group. The estimated rate of CRS
after maternal rubella in the first trimester is 80% [South and
Sever, 1985]. Hence, the power of our cohort to detect differences in rates of CRS exceeds 90%. This study offers further
reassurance with respect to fetal safety for the thousands of

women who are vaccinated against rubella every year not

realizing they are pregnant.

REFERENCES
Bart SW, Stetler HC, Preblud SR, Williams NM, Orenstein WA, Bart KJ,
Hinman AR, Herrmann KL. 1985. Fetal risk associated with rubella
vaccine: An update. Rev Infect Dis 7(Suppl 1):S95S102.
Bolognese RJ, Corson SL, Fuccillo DA. 1973. Evaluation of possible
transplacental infection with rubella vaccination during pregnancy.
Am J Obstet Gynecol 117:939941.
Burgess MA. 1990. Rubella vaccination just before or during pregnancy.
Med J Austral 152:507508.
Canadian Pharmaceutical Association. 2003. Compendium of pharmaceuticals and specialties. 1354p. Ottawa, Ontario.
Centers for Disease Control (CDC). 1989. Rubella vaccination during
pregnancyUnited States, 19711988. MMWR 38:289293.
Ebbin AJ, Wilson MG, Chandor SB, Wehrle PF. 1973. Inadvertent rubella
immunization in pregnancy. Am J Obstet Gynecol 117:505512.
Enders G. 1985. Rubella antibody titers in vaccinated and nonvaccinated
women and results of vaccination during pregnancy. Rev Infect Dis
7(Suppl 1):S103S107.
Fleet WF, Benz EW, Karzon DT. 1974. Fetal consequences of maternal
rubella immunization. JAMA 227:621627.
Hayden GF, Herrmann KL, Buimovici-Klein E, Weiss KE, Nieburg PI,
Mitchell JE. 1980. Subclinical congenital rubella infection associated
with maternal rubella vaccination in early pregnancy. J Pediatr 96:869
872.
Koren G. 1997. Bias against the null hypothesis in maternal-fetal
pharmacology and toxicology. Clin Pharmacol Ther 62:15.
Koren G, Pastuszak AP. 1990. Prevention of unnecessary pregnancy
terminations by counselling women on drug, chemical, and radiation
exposure during the first trimester. Teratology 41:657661.
Marden PM, Smith DW, McDonald MJ. 1964. Congenital anomalies in the
newborn, including minor variations. J Pediatr 64:357371.
Phillips CA, Maeck JVS, Rogers WA, Savel H. 1970. Intrauterine rubella
infection following immunization with rubella vaccine. JAMA 213:624
625.
Preblud SR, Williams NM. 1985. Fetal risk associated with rubella vaccine:
Implications for vaccination of susceptible women. Obstet Gynecol
66:121123.
South MA, Sever JL. 1985. Teratogen update: The congenital rubella
syndrome. Teratol 31:297307.
Tookey PA, Jones G, Miller BH, Peckham CS. 1991. Rubella vaccination in
pregnancy. CDR (Lond Engl Rev) 1:R86R88.
Vaheri A, Vesikari T, Oker-Blom N, Seppala M, Parkmar PD, Veronelli J,
Robbins FC. 1972. Isolation of attenuated rubella-vaccine virus from
human products of conception and uterine cervix. N Engl J Med
286:10711074.
Wyll SA, Herrmann KL. 1973. Inadvertent rubella vaccination of pregnant
women. Fetal risk in 215 cases. JAMA 225:14721476.