Content uploaded by Pierre-Yves François Robillard
Author content
All content in this area was uploaded by Pierre-Yves François Robillard
Content may be subject to copyright.
Available via license: CC BY
Content may be subject to copyright.
Multidisciplinary Prospective Study of
Mother-to-Child Chikungunya Virus Infections
on the Island of La Re
´
union
Patrick Ge
´
rardin
1,2
, Georges Barau
3
, Alain Michault
4
, Marc Bintner
5
, Hanitra Randrianaivo
6
, Ghassan Choker
1
,
Yann Lenglet
3
, Yasmina Touret
3
, Anne Bouveret
3
, Philippe Grivard
4
, Karin Le Roux
4
,Se
´
verine Blanc
5
, Isabelle Schuffenecker
7
,
The
´
re
`
se Couderc
8,9
, Fernando Arenzana-Seisdedos
10,11
, Marc Lecuit
8,9,12}*
, Pierre-Yves Robillard
1}
1 Neonatal and Pediatric Intensive Care Unit, Po
ˆ
le Me
`
re-Enfant, Groupe Hospitalier Sud-Re
´
union, Saint-Pierre, La Re
´
union, France, 2 Center of Clinical Investigation–Clinical
Epidemiology (CIC-EC, INSERM), Saint-Pierre, La Re
´
union, France, 3 Department of Gynecology and Obstetrics, Po
ˆ
le Me
`
re-Enfant, Groupe Hospitalier Sud-Re
´
union, Saint-
Pierre, La Re
´
union, France, 4 Department of Microbiology, Po
ˆ
le des Sciences Biologiques, Groupe Hospitalier Sud-Re
´
union, Saint-Pierre, La Re
´
union, France, 5 Department of
Neuroradiology, Po
ˆ
le de Radiologie, Groupe Hospitalier Sud-Re
´
union, Saint-Pierre, La Re
´
union, France, 6 Fetal Medicine and Fetopathology Unit, Po
ˆ
le Me
`
re-Enfant, Groupe
Hospitalier Sud-Re
´
union, Saint-Pierre, La Re
´
union, France, 7 National Reference Center for Arboviruses, Institut Pasteur, Lyon, France, 8 Institut Pasteur, Microorganisms and
Host Barriers Group, F-75015, Paris, France, 9 Inserm, Equipe Avenir U604, F-75015, Paris, France, 10 Institut Pasteur, Molecular Viral Pathogenesis Unit, F-75015, Paris, France,
11 Inserm U819, F-75 015, Paris, France, 12 Department of Infectious Diseases and Tropical Medicine, Necker-Pasteur Center for Infectious Diseases, Ho
ˆ
pital Necker-Enfants
Malades, Assistance Publique-Ho
ˆ
pitaux de Paris, F-75015, Paris, France
Funding: This study was funded in
part by the Institut Pasteur, the
Agence Nationale pour la Recherche
and INSERM. ML is a recipient of an
INSERM interface contract. The funders
played no role in study design, data
collection and analysis, decision to
publish, or preparation of the
manuscript.
Competing Interests: The authors
have declared that no competing
interests exist.
Academic Editor: Jean-Paul Chretien,
United States Department of Defense,
United States of America
Citation: Ge
´
rardin P, Barau G, Michault
A, Bintner M, Randrianaivo H, et al.
(2008) Multidisciplinary prospective
study of mother-to-child chikungunya
virus infections on the Island of La
Re
´
union. PLoS Med 5(3): e60. doi:10.
1371/journal.pmed.0050060
Received: August 13, 2007
Accepted: January 24, 2008
Published: March 18, 2008
Copyright: Ó 2008 Ge
´
rardin et al. This
is an open-access article distributed
under the terms of the Creative
Commons Attribution License, which
permits unrestricted use, distribution,
and reproduction in any medium,
provided the original author and
source are credited.
Abbreviations: ADC, apparent
diffusion coefficient; APFD,
antepartum fetal death; CHIKV,
chikungunya virus; CI, confidence
interval; CNS, central nervous system;
CSF, cerebrospinal fluid; DHF, dengue
hemorrhagic fever; DIC, disseminated
intravascular coagulation; DWI,
diffusion-weighted imaging; GHSR,
Groupe Hospitalier Sud-Re
´
union;
m[number], month post-outbreak; IQR,
interquartile range; NICU, neonatal
intensive care unit; T1WI, T1-weighted
imaging
* To whom correspondence should be
addressed. E-mail: mlecuit@pasteur.fr
} These authors are joint senior
authors on this work.
ABSTRACT
Background
An outbreak of chikungunya virus affected over one-third of the population of La Re
´
union
Island between March 2005 and December 2006. In June 2005, we identified the first case of
mother-to-child chikungunya virus transmission at the Groupe Hospitalier Sud-Re
´
union level-3
maternity department. The goal of this prospective study was to characterize the
epidemiological, clinical, biological, and radiological features and outcomes of all the cases
of vertically transmitted chikungunya infections recorded at our institution during this
outbreak.
Methods and Findings
Over 22 mo, 7,504 women delivered 7,629 viable neonates; 678 (9.0%) of these parturient
women were infected (positive RT-PCR or IgM serology) during antepartum, and 61 (0.8%) in
pre- or intrapartum. With the exception of three early fetal deaths, vertical transmission was
exclusively observed in near-term deliveries (median duration of gestation: 38 wk, range 35–40
wk) in the context of intrapartum viremia (19 cases of vertical transmission out of 39 women
with intrapartum viremia, prevalence rate 0.25%, vertical transmission rate 48.7%). Cesarean
section had no protective effect on transmission. All infected neonates were asymptomatic at
birth, and median onset of neonatal disease was 4 d (range 3–7 d). Pain, prostration, and fever
were present in 100% of cases and thrombocytopenia in 89%. Severe illness was observed in
ten cases (52.6%) and mainly consisted of encephalopathy (n ¼ 9; 90%). These nine children had
pathologic MRI findings (brain swelling, n ¼ 9; cerebral hemorrhages, n ¼ 2), and four evolved
towards persistent disabilities.
Conclusions
Mother-to-child chikungunya virus transmission is frequent in the context of intrapartum
maternal viremia, and often leads to severe neonatal infection. Chikungunya represents a
substantial risk for neonates born to viremic parturients that should be taken into account by
clinicians and public health authorities in the event of a chikungunya outbreak.
The Editors’ Summary of this article follows the references.
PLoS Medicine | www.plosmedicine.org March 2008 | Volume 5 | Issue 3 | e600413
P
L
o
S
MEDICINE
Introduction
The chikungunya virus (CHIKV) is an enveloped, positive-
strand RNA alphavirus belonging to the Togaviridae family
and transmitted by Aedes mosquito bites [1]. It causes a
dengue-like illness, characteriz ed by feve r, rash, painful
myalgia, and arthralgia, and sometimes arthritis [2]. It was
first isolated by R.W. Ross in 1952 in the Newala district of
Tanzania [3]. Its current geographic distribution covers sub-
Saharan Africa, Southeast Asia, India, and the Western Pacific
where numerous outbreaks have been reported [4–8]. In these
areas, upsurges of re-emergence occur at intervals of 7 to 20
years [9].
Since the end o f 2004 , CHIKV has emerged in the
Southwestern Indian Ocean islands. Between January and
March 2005, over 5,000 cases were reported in the Comoros.
Later in 2005, the virus spread to other islands, including
Mayotte, Seychelles, La Re´union, and Mauritius [9]. In La
Re´union Island, a French overseas department (total pop-
ulation: 787,836), the first declared case was observed in
Saint-Pierre (southern area of the island) in the beginning of
March 2005 among people returning from the Comoros [10].
The transmission was moderate until the rainy season, which
started in December in 2005 and was associated with an
epidemic of unprecedented magnitude (300,000 cumulative
cases on December 30, 2006), with a peak incidence reached
on the fifth week of 2006 (over 45,000 cases). No other known
arboviral disease was associated with this chikungunya out-
break.
Aedes albopictus was identified as the only vector of a
principally urban transmission in La Re´union [11]. During
this outbreak, severe or complicated forms of chikungunya
were reported in adult patients, including encephalopathy
and hemorrhagic fever, which frequently occurred in the
context of chronic diseases or underlying conditions such as
diabetes mellitus, chronic obstructive pulmonary disease,
ischemic heart disease, chronic renal failure, or alcoholic
hepatopathy [10,12].
In June 2005, we identified the first case of mother-to-child
chikungunya virus transmission [13]. We thus conducted a
prospective study in order to characterize the epidemiolog-
ical, clinical, biological, and radiological features and out-
comes of all the cases of mo ther-to-child chikungunya
infections recorded at our institution during this outbreak.
Methods
Study Location and Participants
Our prospective study took place in the level-3 public
maternity department of the Groupe Hospitalier Sud -
Re´union (GHSR), the largest hospital on the island (1,300
beds, standard care comparable to that available in Europe,
4,300 deliveries per year for a population of 300,000
inhabitants; 80% of the births within the area, the remaining
taking place in a private level-1 maternity department). The
level-3 maternity department of the GHSR is well insulated
and air-conditioned, and at the peak of the chikungunya
outbreak, blood-derived products were imported from a
chikungunya-free area, minimizing the risk of nosocomial
CHIKV transmission, whether by mosquito bite or blood-
derived products.
We enrolled all parturient women and their offspring
admitted at the maternity department between 1 June 2005
(date of the first chikungunya infection observed in the
course of a pregnancy) and 30 December 2006 (date of
delivery of the last pregnant woman infected with chikungu-
nya). For the use of the data, oral consent was obtained from
each patient or a first-degree relative, as the investigations
were carried out under the standard care procedure for this
new disease, in accordance with the recommendations of the
Committee for Clinical Research of the GHSR. In France,
written consent is mandatory only if the medical treatments
or the products used are not standard for the diagnosis,
treatment, or monitoring (art. 88-II, law 2004–806, Journal
Officiel, 08/11/2004; art. 31-I, law 2006–450, Journal Officiel, 04/
19/2006). The information was given in French or in Creole
with the help of a translator when necessary.
Data Collection and Screening
Midwives collected full obstetric history from mandatory
maternity booklets and additional questioning in the frame-
work of our daily epidemiologic perinatal survey [14].
As the positive predictive value for chikungunya infection
of the association of fever with rash or arthralgia during the
outbreak was higher than 95% [15], and the rate of clinically
silent cases below 5% (Ge´rardin et al., unpublished data),
serological screening for chikungunya was performed in all
pregnant women who had presented with these clinical signs
during the course of their pregnancy but were not diagnosed
with chikungunya infection prior to their pregnancy. The
delay for IgM seroconversion was 5 to 7 d after the onset of
symptoms, whereas that for IgG was up to 15 d [16]. CHIKV-
specific IgM antibodies were detected by capture ELISA [17]
and CHIKV-specific IgG antibodies by ELISA [18] developed
at the Centre National de Re´fe´rence des Arbovirus (Institut
Pasteur, Lyon, France) and automated with an EtiMax 3000
apparatus (DiaSorin, Italy). Each serum sample was assayed in
a well coated with culture supernatant of a CHIKV-infected
cell line (wAg) and in a well coated with culture supernatant
of the corresponding uninfected cell line (woAg). A reference
serum (ref serum) was tested in triplicate on each 96-well
microplate. The results were expressed in arbitrary units to
normalize the inter-assay variability and were calculated as
follows: (OD serum sample wAg DO serum sample woAg) /
[(R
13
(OD ref serum wAg OD ref serum woA) / 3] 3 100. The
cut-off values for IgMs and IgGs were determined from a
series of 30 negative sera collected in La Re´union Island
before the epidemic started. Positive and negative control
sera were added on each 96-well microplate.
For symptomatic women, whether presenting for exami-
nation at the outpatient clinic or for hospitalization at the
maternity ward, a one-step TaqMan real time quantitative
RT-PCR was performed in serum samples using the Light
Cycler 2.0 system (Roche Diagnostics). Oligonucleotide
primers amplified a conserved coding region of glycoprotein
E1 [19]. A positive homologous internal control was included
for monitoring RNA extraction and detecting false negatives.
A synthetic RNA was used as an external standard for CHIKV
quantification. The assay sensitivity for serum specimen was
350 genome copies per milliliter [20]. The window of positive
real-time quantitative RT-PCR ranged from the day of the
onset of symptoms (day 0) to day 5.
The timing of chikungunya maternal infection was deter-
mined after delivery, based on clinical criteria as well as RT-
PLoS Medicine | www.plosmedicine.org March 2008 | Volume 5 | Issue 3 | e600414
Maternal-Fetal Chikungunya Virus Infection
PCR and IgM/IgG serology results (see Table 1). In the absence
of evidence of earlier infection, maternal chikungunya was
classified as antepartum in 678 women whose clinical signs
had occurred between conception and the week preceding
labor, and diagnosed by RT-PCR performed prospectively for
outpatients, or retrospectively at delivery by detection of IgM
antibodies. Maternal chikungunya was classified as prepartum
in 22 women with symptoms lasting between day 7 and day
3 before delivery and diagnosed by RT-PCR (or IgM
seroconversion when not available), and as intrapartum in
39 women with symptoms between day 2 and day 2 around
delivery and concomitant positive RT-PCR (or IgM serocon-
version when not available).
Exposed neonates were screened by IgM and IgG assays at
days 0, 7, and 15. For 57 out of 62 neonates born from a
mother viremic in pre- or intrapartum, a RT-PCR test was
also performed on serum at day 0 and/or during the first week
of life upon clinical indication. The RT-PCR and serological
results for the diagnosis of chikungunya are provided for 591
out of 749 neonates in Table 2.
Clinical features associated with neonatal infections were
collected during the stay in the neonatal department (28 beds,
ten for neonatal intensive care unit [NICU]; 350 admissions
per year). For neonates, the chikungunya was considered as a
vertical mother-to-child transmission when symptoms oc-
curred during the first week of life, in the absence of evidence
of mosquito bite. For mothers and neonates, cases not
confirmed by RT-PCR or IgM serology were excluded.
Clinical variables included the mode of delivery, fetal heart
rate, gestational age, birth weight, 5-min Apgar score, delay
between birth and onset of symptoms (defining the incuba-
tion period), fever, cutaneous and rheumatologic signs (rash,
petechiae, and joint edema), hematological parameters (total
blood cell count and hemoglobin), and blood biochemical
parameters (sodium, glucose, calcium, urea, creatinine, and
liver enzymes). In case of a severe thrombocytopenia (platelet
count , 100,000/mm
3
), blood coagulation tests (prothrombin
time, partial thromboplastin time, fibrinogen, and D-dimers)
were performed to diagnose disseminated intravascular
coagulation (DIC) syndrome.
The brain MRI protocol included T1-weighted imaging
(T1W I) before and after the intravenous infusion of a
gadolinium-based contrast agent, T2-weighted imaging
(T2WI), and diffusion-weighted imaging (DWI) with apparent
diffusion coefficient (ADC) maps. Neurological follow-up of
infected newborns after discharge from the neonatal depart-
ment was performed using a standardized neurological test
[21].
Statistical Analyses
Monthly cumulative incidence rates, i.e., attack rates, were
measured as the ratio of the pregnant women newly infected
per month divided by the sum of the women delivered during
Table 1. Serological and/or RT-PCR Evidence for Chikungunya Diagnosis in Ante-, Pre-, and Intrapartum Infections/Exposures, Groupe
Hospitalier Sud-Re
´
union, Saint-Pierre, La Re
´
union, France, 2005–2006 Outbreak: Mothers
Assay Antepartum Infections n ¼ 678 Prepartum Infections n ¼ 22 Intrapartum Infection n ¼ 39
Maternity
Department
Ambulatory or
Outpatient Care
Maternity
Department
Ambulatory or
Outpatient Care
Maternity
Department
Ambulatory or
Outpatient Care
IgM and IgG NA, PCR þ —29 3— 6—
IgM , IgG NA, PCR þ — — 8 — 24 —
IgM þ,IgGþ, PCR þ —7 —— ——
IgM þ, IgG NA, PCR þ —14 3— 2—
IgM þ, IgG and PCR NA 388 166 5 — 4 —
IgM þ, IgG and PCR —— 3
a
———
IgM and IgG þ, PCR NA (retrospectively) 74 — — — — 3
b
Total n ¼ 739.
a
RT-PCR was performed at delivery for infected mothers referred late in the course of their infection.
b
Three mothers of infected neonates assessed retrospectively.
NA, not available.
doi:10.1371/journal.pmed.0050060.t001
Table 2. Serological and/or RT-PCR Evidence for Chikungunya
Diagnosis in Ante-, Pre-, and Intrapartum Infections/Exposures,
Groupe Hospitalier Sud-Re
´
union, Saint-Pierre, La Re
´
union,
France, 2005–2006 Outbreak: Viable Neonates
Viable Neonates
a
Antepartum
Exposures
n ¼ 687
b
Prepartum
Exposures
n ¼ 22
Intrapartum
Exposures
n ¼ 40
b
IgM and IgG NA, PCR þ —— 1
c
IgM þ, IgG NA, PCR þ ——15
c
IgM þ, IgG NA, PCR —— 3
c,d
IgM ,IgGþ, PCR NA 223 — —
IgM ,IgGþ, PCR 33 — —
IgM , IgG NA, PCR 138 12 21
IgM , IgG and PCR NA 102 4 —
IgM ,IgG, PCR 31 — —
IgM, IgG, and PCR NA 157 — —
IgM NA, IgG , PCR NA 3 1 —
IgM and IgG NA / PCR —5—
Total n ¼ 749.
a
Neonates were born after 22 wk of amenorrhea (excluding early fetal deaths).
b
Ten dizygous twins.
c
Infected neonates.
d
Three neonates tested late in the course of their infection, and thus RT-PCR negative but
IgM positive.
NA, not available.
doi:10.1371/journal.pmed.0050060.t002
PLoS Medicine | www.plosmedicine.org March 2008 | Volume 5 | Issue 3 | e600415
Maternal-Fetal Chikungunya Virus Infection
the same month plus those anticipated to deliver in the eight
following months. Monthly prevalence rates were calculated
for the parturient women as the ratio of the number of
parturients infected during their pregnancy and delivered in
the month over the number of parturients in the given
month.
In or der to det ermi ne the obstetrical and neonatal
characteristics of mother-to-child transmission of CHIKV, a
case-control study was conducted, after exclusion of early
antepartum fetal deaths (APFDs) before 22 wk. Clinical and
biological continuous parameters were compared with the
Mann-Whitney test. The rates of fetal heart decelerations,
cesareans sections (C-sections), and neonatal asphyxia (de-
fined as 5-min Apgar score , 7) for the cases of vertical
transmission of CHIKV and for a randomly selected control
group of uninfected neonates of the same gestational age
range (born from uninfected mothers and hospitalized at the
same time in the neonatal care unit) were compared using
Chi-square or Fisher exact tests, when appropriate. Two
controls per case were included. All analyses were computed
in Stata (Stata Statistical Software: release 9; StataCorp. 2005).
A p-value , 0.05 was considered statistically significant.
Results
During this 22-mo long survey (March 2005 to December
2006), 7,504 consecutive women delivered 7,629 viable neo-
nates at the level-3 GHSR maternity department whilst about
2,000 births occurred in the level-1 maternity department.
The monthly evo lution of the cumulative incidence of
maternal chikungunya infections during pregnancy and that
of neonatal cases observed in the GHSR level-3 maternity are
presented in Figure 1. The first reported antepartum case
occurred in May 2005 (third month after the beginning of the
outbreak, m3) and the last in June 2006 (m16). During the first
9 mo of the outbreak (cold season in the Southern hemi-
sphere), the attack rate among pregnant women was below
1% and the prevalence rate among parturient women was
below 5%, owing to a sporadic transmission (fewer than ten
cases per week). At m10, the hot and rainy season (austral
summer) started, and the attack rates increased sharply
during m10–m12 in the general population (45,000 new cases
during the first week of February 2006). For pregnant women,
the incidence peaked in January 2006 (m11) with an attack
rate of 8.3% (95% confidence interval [CI] 7.4%–9.3%).
Among parturient women, the peak of prevalence was
Figure 1. Monthly Evolution of Neonatal, Maternal Pre- and Intrapartum, and Antepartum Chikungunya Cases between March 2005 (m1) and June 2006
(m16), First Sud-Re
´
union Outbreak, in Saint-Pierre
doi:10.1371/journal.pmed.0050060.g001
PLoS Medicine | www.plosmedicine.org March 2008 | Volume 5 | Issue 3 | e600416
Maternal-Fetal Chikungunya Virus Infection
delayed to m15 with a 27.5% reported rate. Between m13 and
m16 the attack rates decreased sharply to 0.4% (95% CI
0.15%–0.6%) and then prevalence rates felt dramatically to
reach only 0.4% in December 2006 (m22), marking the end of
our survey.
In total, of the 7,504 parturient women, 739 (9.8%)
reported a history of chikungunya during pregnancy, includ-
ing 678 (9.0%) in the antepartum period (i.e., more than a
week before delivery), and 61 (0.8%) in pre- (n ¼ 22) or
intrapartum (n ¼ 39) (i.e., symptoms between day 7 and day
3, or day 2 and day 2 around delivery, respectively) (Table
1). Chikungunya maternal infection (i.e., fetal exposure to
maternal blood-borne chikungunya) distribution was uni-
form throughout pregnancy (median onset of maternal
infection: 25 wk; interquartile range [IQR] 16–33; range 0–
41; mode 36), and no significant breakdown or peak in
infection during pregnancy was observed. During the same
period, fewer than 20 cases of antepartum maternal infection,
and no pre- or intrapartum maternal or neonatal infection,
were reported in the level-1 private maternity department.
The prevalence rate of APFD after 22 wk for the Sud-
Re´union area throughout the chikungunya outbreak did not
differ from previous annual rates available since 2001 (0.9%
between March 2005 and December 2006, 0.5% in 2001, 0.7%
in 2002, 1% in 2003, and 1.2% in 2004). Among the 678
women infected antepartum and whose pregnancy was
monitored from the second trimester, nine APFD were
reported after 22 wk, and none was attributable to CHIKV
infection (negative CHIKV RT-PCR for amniotic fluid, fetal
brain, and serum). Among the seven early APFDs that were
reported before 22 wk, only three were attributable to
CHIKV infection: the three pregnant women were viremic
(positive serum CHIKV RT-PCR) at the onset of symptoms (12
wk 4d, 15 wk, and 15 wk 5d), and APFD was observed around
two weeks later [22]. For these three APFDs, the amniotic
fluid collected by amniocenteses before fetal demise was RT-
PCR positive. CHIKV RNA was detected in the placenta and
in the fetal brain for two. These three women were no longer
viremic at the time of miscarriage, excluding a postmortem
contamination of the fetus from the maternal blood. Of the
four remaining early APFDs, CHIKV RNA was not amplified
from fetal samples.
None of the children of the 678 pregnant women infected
antepartum had dete ctable IgM at birth (Table 2), and
mother-transferred IgG, surveyed in 70 infants, cleared
progressively: 3% of the infants were IgG negative at 3 mo
of age, 43% at 6 mo, and 81% at 9 mo. On follow-up, the
absence of IgM at 3 months in these children ensured that
they had not been infected.
Most pre- and intrapartum maternal cases were diagnosed
by RT-PCR, and only three prepartum-in fected w omen
referred late in the course of the infection were RT-PCR
negative (Table 1). Among the 61 women who presented with
ongoing chikungunya infection in the setting of delivery (22
in pre- and 39 in intrapartum) (Table 1), 19—all with an
intrapartum infection—transmitted the chikungunya to their
offspring (vertical mother-to-child transmission rate for
intrapartum infections: 19/39, 48.7%) (Table 2). All mother-
Table 3. Admission Characteristics of 19 Neonatal Cases of Chikungunya Mother-to-Child Infection and of 38 Unexposed Controls
Treated in the Neonatal Care Unit of Groupe Hospitalier Sud-Re
´
union, Saint-Pierre, La Re
´
union, France, 2005–2006 Outbreak
Parameters Category Physiological
Ranges
Cases, n ¼ 19 Controls, n ¼ 38 p-Value
n or
Median
Percentage
or IQR
n or
Median
Percentage
or IQR
Mode of delivery Vaginal — 9 (47.4) 19 (50.0) —
Planned C-section — 1 (5.2) 6 (15.8) 0.644
a
Emergency C-section — 9 (47.4) 13 (34.2) 0.731
Fetal heart rate monitoring — — — — — — 0.015
Normal — 5 (26.3) 23 (60.5) —
Decelerations — 14 (73.7) 15 (39.5) —
Gestational age (weeks of amenorrhea) — — 38 (35–40) 38 (34–41) 0.715
Birth weight (g) — — 3,010 (2,675–3,345) 2,870 (2,333–2,993) 0.078
5-min Apgar score — — — — — — 0.304
7 — 17 (89.5) 29 (76.3) —
, 7 — 2 (10.5) 9 (23.7) —
Highest skin temperature (8C) — — 38.8 (38.1–39.1) 37.2 (37.0–37.3) , 0.001
Length of stay (d) — — 15 (12–21) 4.5 (2–12) , 0.001
Hematology Lymphocyte counts (/mm
3)
(2,000–17,000) 800 (600-1550) 3,400 (2,500–4,500) , 0.001
Hemoglobin level (g/dl) (12.0–20.0) 14.6 (13.0–17.8) 17.9 (15.9–19.3) 0.028
Platelet counts (3 1,000/mm
3
) (150–400) 52 (21.5–106.5) 241 (175–304) , 0.001
Biochemistry Prothrombin time (%) (150–400) 63 (42–79) 57 (51 – 65) 0.845
Serum sodium (mM) (130–146) 139 (135.5–143) 141 (139–143) 0.431
Serum glucose (mM) (2.8–4.4) 4.0 (3.4–4.6) 3.6 (3.1–4.1) 0.087
Serum urea (mM) (1.0–4.2) 4.0 (3.4–6.2) 3.4 (2.7–4.8) 0.148
Serum creatinine (lM) (20–50) 46 (39.5–54) 50 (44–59) 0.865
Serum calcium (mM) (2.0–2.7) 2.1 (1.8 – 2.2) 2.5 (2.3–2.6) , 0.001
Serum AST (IU/l) (30–110) 103.5 (75–123.2) 30 (27–35) 0.014
Comparisons were made using the Chi square, Fisher exact, or Mann-Whitney test as appropriate.
a
p ¼ 0.925 when combining the indications of C-sections (Chi square test).
doi:10.1371/journal.pmed.0050060.t003
PLoS Medicine | www.plosmedicine.org March 2008 | Volume 5 | Issue 3 | e600417
Maternal-Fetal Chikungunya Virus Infection
to-child transmissions occurred in the context of near-term
deliveries (median length of gestation: 38 wk, range 35–40
wk). During the labor of the 61 women experiencing viremia
around term, none received blood-derived product before
the section of the umbilical cord. Forty-six fetuses (75%)
exhibited deep spikes or late decelerations on fetal heart
monitoring. These abnormalities were seen in the cases of
vertical transmission (14/19) as well as in its absence (32/42),
and were more likely to occur in CHIKV-infected women
than that from uninfected control neonates (see Methods)
(73.7% versus 39.5%, p ¼ 0.01, Table 3). Of the 61 women
experiencing a CHIKV infection in the setting of delivery, the
rate of C-section was 42.6%, of which 69.2% were performed
because of acute fetal distress. This rate significantly
exceeded the 17.4% overall C-section rate observed in our
center (p 0.001) [14]. However, it was not significantly
different than that of randomly selected parturients whose
neonates were hospitalized at the same time in the neonatal
care unit (Table 3). Importantly, C-section had no influence
on mother-to-child viral transmission, either for intrapartum
infections (C-section rates in infected versus uninfected
neonates: 48.7% versus 52.9%, NS), or globally for pre-/
intrapartum infections (unpublished data).
The viral load in the placentas of seven of the 19
transmitters (mean 6 standard deviation [SD]: 42,000 6
20,167 copies/mg of tissue) was significantly higher than that
in 13 placentas of the nontransmitters (mean 6 SD: 10,742 6
8,182 copies/mg, Mann Whitney test, p ¼ 0.021). Among the 19
transmitters, one gave birth to dizygous twins: one neonate
remained uninfected, whereas the other became infected.
Of the 19 neonates who developed a vertical chikungunya
infection in the setting of maternal viremia, all were exposed
in intrapartum and none was symptomatic at birth nor had a
history of mosquito bite. In none of the 16 neonate s
diagnosed by RT-PCR (mean viral load 250 million copies/
ml of plasma) was the CHIKV genome detected at day 1 (viral
load , 350 copies/ml). The median onset of neonatal disease
(defining the incubation time) was 4 d after birth (range 3–7
d). Among the 19 infected neonates, ten were breast-fed.
Maternal milk, tested for 20/33 breast-feeding viremic
women, was always RT-PCR negative. No infection wa s
reported over the same time among hospitalized neonates
born from non-viremic mothers.
In our series, the main clinical features at presentation
were fever, poor feeding, and pain, observed in all infected
infants (100%) and evidenced by need for constant analgesic
treatment and enteric feeding. Rheumatic and cutaneous
signs also constituted major associated clinical manifesta-
tions: distal joint edema (15/19, 78.9%), petechiae (9/19, 47.3
%), or polymorphous rubella-like (10/19, 52.6%) or roseola-
like exanthema (7/19, 36.8%).
The most frequent physiological abnormality was throm-
bocytopenia (17/19, 89.4%), associated with a mild elongation
of prothrombin time (n ¼ 6) and DIC (n ¼ 4). Severe
thrombocytopenia (9/19, 47.3 %) led to empiric administra-
tion of hydrocortisone (n ¼ 8), gamma globulins (n ¼ 1), or
both (n ¼ 2), and platelet infusions (n ¼ 4) with the aim of
avoiding massive hemorrhages, known to be life-threatening
in the context of dengue hemorrhagic fever (DHF). Other
biological abnormalities included lymphopenia (13/19,
68.4%), which preceded the onset of clinical signs in half of
the cases, a mild increase of AST level (10/19, 52.6%), and a
moderate to severe hypocalcemia (9/19, 47.3 %).
The admission characteristics of the 19 neonates treated at
our center for vertical chikungunya infection are presented
and compared to unexposed controls in Table 3.
Severe neonatal disease was observed in ten cases (52.6%)
and mainly consisted of encephalopathy (n ¼ 9), or hemor-
rhagic fever (n ¼ 1). Cerebrospinal fluid (CSF), collected for
the nine cases with encephalopathy, was RT-PCR positive for
CHIKV in five cases (mean viral load 184,000 copies/ml of
CSF), but normal for chemistry and cytology (except for three
cases with more than 3,500 RBCs/mm
3
, one of which was in
the context of DIC). Of the ten severe cases, eight required
mechanical ventilation (median duration: 7 d). Shock/hypo-
volemia was observed in six cases; each displayed a hyper-
kinetic profile on echocardiography, and four required the
use of vasoactive amines (median duration 2 d). DIC, reported
in four cases, was complicated by transient brain hemor-
rhages in two—namely, scattered parenchymal petechiae in
both hemispheres with additional bleeding, cerebellar hem-
atoma in one, and hematemesis in the other. Indicators of
severity at admission were a normal skin temperature (p ¼
0.008), a low prothrombin time (p ¼ 0.002), and a low platelet
count (p ¼ 0.035) (presented in detail in Table 4).
The most distinctive MRI abnormalities observed in the
course of neonatal chikungunya encephalopathy are shown in
Figure 2. At the acute phase stage (6–15 d after onset of
Table 4. Factors Associated with Severe Disease in the 19 Neonates Treated in the Neonatal Intensive Care Unit of Groupe Hospitalier
Sud-Re
´
union, Saint-Pierre, La Re
´
union, France, for a Chikungunya Mother-to-Child Infection, 2005–2006 Outbreak
Parameters Physiological
Ranges
Severe Cases, n ¼ 10 Mild Cases, n ¼ 9 p-Value
Median IQR Median IQR
Gestational age at birth (weeks of amenorrhea) — 37.5 (36.3–38.0) 39.0 (38.0–39.0) 0.017
Birth weight (g) — 2,675 (2,343–3,033) 3,320 (3,010–3,760) 0.010
Highest skin temperature (8C) — 38.1 (38.0–38.8) 39.1 (38.8–39.6) 0.008
Length of stay (days) — 22 (18–31) 12 (6–13) , 0.001
Lymphocyte counts (/mm
3
)
a
(2,000–17,000) 700 (600-1600) 900 (600-1400) 0.666
Platelet counts (x 1000/mm
3
) (2,000–17,000) 34.5 (15,75–51,0) 90 (60–132) 0.035
Prothrombin Time (%) (70–100) 41 (36–54) 74 (72–100) 0.002
a
Lymphocyte counts were rounded to the nearest hundred
doi:10.1371/journal.pmed.0050060.t004
PLoS Medicine | www.plosmedicine.org March 2008 | Volume 5 | Issue 3 | e600418
Maternal-Fetal Chikungunya Virus Infection
disease), MRI revealed unremarkable features on T1WI and
T2WI, but several scattered and sometimes hyperintense
signals of the supratentorial white matter were observed
(Figure 2A), involving the corpus callosum and the frontal,
parietal, and temporal lobes on DWI evocative of an early
cytotoxic edema. Similar images were observed in the six
cases for which MRI was obtained at the acute phase, and
were associated with a marked reduction of the ADC (Figure
2B), compatible with parenchymal ischemia. Early evolution
in the subacute phase (15–45 d after onset of disease) showed
that the areas of hyperintense signals were replaced by areas
of very low intensity on DWI (Figure 2C) in favor of an
evolution into a vasogenic edema. These images were
observed in all nine cases of encephalopathy and were
associated with a concomitant increase of the ADC (Figure
2D) show ing the reperfusion of low-output areas. They
regressed in seven neonates but evolved towards cavitations
and subcortical atrophy in two (unpublished data). No infants
died, even those with severe shock or massive hemorrhage.
For the nine neonates with encephalopathy and brain
swelling images, neurological sequelae were assessed upon
discharge and throughout a 16- to 24-month follow-up. Four
evolved towards persistent disabilities: one developed cere-
bral palsy with ataxia and blindness following extensive white
matter degeneration for which a CACH/VWM (childhood
ataxia with central nervous system hypomyelination/leucoen-
cephalopathy with vanishing white matter) syndrome was
ruled out (no mutation on EIF2B); three had ocular and
behavioral or postural deficiencies (dysconjugate gaze, n ¼ 3;
language delay, n ¼ 2; axial hypotonia, n ¼ 1). These four
children had multiple seizures during their original neonatal
hospitalization and one of them still requires anticonvulsants
Figure 2. Representative MRI Findings in Neonatal Chikungunya Encephalopathy Cases Observed during the First Chikungunya Outbreak in Saint-
Pierre, La Re
´
union, June 2005 to December 2006
(A) Day 6 (child A), scattered hyperintensity signals of the white matter involving the corpus callosum, the frontal and parietal lobes on diffusion-
weighted imaging (DWI).
(B) Day 6 (child A), scattered reduction of the Apparent Diffusion Coefficient (ADC) of the white matter involving the corpus callosum, the frontal and
parietal lobes on ADC mapping (in blue).
(C) Day 21 (child B), scattered and characteristic hypo-intensity signals of the white matter involving frontal and parietal lobes on DWI (Note that the
corpus callosum remains in hyper-signal).
(D) Day 21 (child B), scattered increase of the ADC of the white matter involving the corpus callosum, the frontal and parietal lobes on ADC mapping (in
red).
doi:10.1371/journal.pmed.0050060.g002
PLoS Medicine | www.plosmedicine.org March 2008 | Volume 5 | Issue 3 | e600419
Maternal-Fetal Chikungunya Virus Infection
at this writing. For the ten other neonatal chikungunya
infections of our series, the clinical status and the brain MRI,
assessed on regular follow-up, were considered normal.
Discussion
Here we report the epidemiological, clinical, biological,
and radiological features and outcomes of maternal-fetal
transmission of CHIKV infection in an outbreak that
occurred on the island of La Re´union, France. We find that
mother-to-child CHIKV transmission is almost exclusively
observed in the context of intrapartum maternal viremia, and
often leads to severe neonatal infection. CHIKV thus
represents a significant risk for neonates born from viremic
parturients that must be taken into account by clinicians and
public health authorities in the event of a chikungunya
outbreak.
Given that the GHSR level-3 maternity department
receives all pregnant women at risk in Sud-Re´ union and that
the level-1 maternity department declared a very low rate of
maternal CHIKV infections because of a lower exposure of its
population [15], and no neonatal infection, one may assume
that the incidence and the prevalence rates calculated in our
survey represent a good approximation of the true burden of
chikungunya, both in pregnant women and in neonates, for
the southern area of La Re´union Island.
We demonstrate here that mother-to-child transmission of
CHIKV is relatively rare, since 2.5% (19/749) of exposed
neonates became infected, although 10% (749/7,629) of the
neonates were exposed during pregnancy, leading to an
overall prevalence of maternal-fetal infections after 22 wk of
0.25% (19/7,629). In contrast, during the delivery period, the
rate of transmission for viremic women was close to 50%,
highlighting the intrapartum period as the critical time for
transmission to the neonate. To our knowledge, the youngest
child ever diagnosed with a chikungunya infection before this
outbreak was a 21-d-old infant who was i nfected by a
mosquito bite [23]. N one of the 19 infected neonates
recorded in this study, all of whom were born to mothers
with an intrapartum infection, presented evidence of
mosquito bites. Moreover, no infection was reported among
concomitantly hospitalized neonates born to mothers with-
out infection, emp has izing the absence of exposure to
mosquito bites or contaminated blood-derived products in
the maternity department. Together, these data provide
evidence that these neonatal infections were the consequence
of mother-to-child transmission complicating intrapartum
maternal viremia. The absence of detectable neonatal viremia
at day 1 of life is consistent with intrapartum maternal–fetal
viral transmission.
Our preliminary investigations on the pathophysiology of
maternal–fetal CHIKV transmission easily ruled out neonatal
contamination through the genital tract, because gastric
aspirations and nasal swabs were RT-PCR negative and C-
sections had no protective effect on viral transmission [24].
Even though the mean viral load of placentas from infected
neonates was significantly higher than that of uninfected
neonates (p ¼ 0.021), immunofluorescence labeling with anti-
chikungunya antibody did not allow the detection of infected
cells in any of these placentas (unpublished data), nor in an
animal model of maternal-fetal chikungunya infection [25]. In
addition, RT-PCR performed on isolated cells obtained from
mechanical dissociation of these placentas was consistently
negative (unpublished data), a result favoring the hypothesis
of placental passive contamination by maternal blood-borne
free virus particles rather than an actual placental infection.
This hypothesis fits our observation that, in contrast to many
in vitro-cultured cell types, the syncytiotrophoblast is not
permissive to CHIKV [25].
Taken together, these observations suggest that the
placental barrier is effective during antepartum in prevent-
ing maternal–fetal CHIKV transmission, as it was docu-
mented in only three cases (3/678, 0.4%) [22]. In contrast, viral
maternal–neonatal transmission is frequently observed in
viremic mothers around the term of pregnancy, when the
highly viremic maternal blood (mean viral load 1.5 million
copies/ml of plasma) can be in contact with placental barrier
breaches resulting from uterine contractions during labor
(vertical transmission rate of 48.7% in neonates exposed
during labor). The higher viral load measured in placentas
from infected neonates is thus most likely a consequence of a
higher maternal viremia, which could therefore be predictive
of the likelihood of transmission. We could not confirm this
hypothesis, because of a lack of concomitant maternal blood
samples corresponding to these placentas.
In areas where the Aedes vector is present, painful arthralgia
complicating a dengue-like disease are strongly evocative of
chikungunya [26]. In neonates, although evidence is limited,
we estimate that painful arthralgia were present in 78%–
100% of our cases, associated with distal joint edema and
persistent prostration. Cutaneous signs, namely polymor-
phous rubelliform and roseoliform exanthema, were common
findings in maternal–fetal neonatal chikungunya infections.
These observations are consistent with previous chikungunya
descriptions that reported possible maculopapular rash in
adults [2,27] and in children [23,28] and that illustrate the
frequency of skin manifestations in human arboviral diseases
[29,30].
A low lymphocyte count was a common finding in neonatal
chikungunya infection (nearly 70% of the cases), justifying
the surveillance of white blood cell counts in neonates in our
maternity hospital during the outbreak. However, as for
dengue fever, lymphopenia is not a marker of severity in
neonatal chikungunya [31]. In contrast, the intensity of
thrombocytopenia, observed in 89% of infected neonates,
was associated with severe neonatal disease and led to the
administration of multiple supportive interventions includ-
ing steroids and gammaglobulins to avoid bleeding compli-
cations, although their benefit have been demonstrated
neither in DHF [32,33] nor in chikungunya. In our series,
severe central nervous system (CNS) hemorrhages, namely
scattered cerebrum-pa renchyma petechiae, were seldom
observed, but always in a clinical context of DIC syndrome,
highlighting the rarity of chikungunya per se as a cause of
hemorrhagic fever [28,34]. These clinical data are consistent
with those reported by Ramful et al., who recently published a
clinical description of the 38 neonatal cases reported to the
local health authorities throughout the outbreak in La
Re´union Island [28]. These investigators reported possible
cardiac involvement, which appears to have been mainly
characterized by coronary artery dilatations (six out of the 16
cases who underwent cardiovascular investigations). How-
ever, given its retrospective design and the fact that it focused
on infected mother–child pairs but not on the whole
PLoS Medicine | www.plosmedicine.org March 2008 | Volume 5 | Issue 3 | e600420
Maternal-Fetal Chikungunya Virus Infection
population of uninfected and infected mothers throughout
pregnancy, this study could not evaluate the modalities of
neonatal transmission nor assess the transmission rate.
Cases of encephalopathy have been reported in adults
during the La Re´ union Island outbreak [12]. These cases were
not associated with CSF, EEG, and MRI abnormalities
(Lemant et al. unpublished data). In contrast, CNS involve-
ment was observed in one-third of our neonatal cases, and
was associated with a massive brain swelling as evidenced by
MRI. This swelling may account for a transiently increased
permeability of the blood–brain barrier without virus-
induced damage of the CNS (RT-PCR–positive CSF but
normal cytology and protein level in CSF samples, and
absence of gadolinium-contrast enhancement), although a
direct virus-induced encephalitis cannot be ruled out. In
neonates and some adults, notably those with underlying
conditions, CHIKV thus appears to exhibit a neurotropism,
to our knowledge not yet described. Other neurotropic
arboviruses include other alphavirus species, such as Eastern
equine encephalitis virus and Venezuelan equine encephalitis
virus, which are both associated with severe encephalitis in
adults, as well as flavivirus family members such as West Nile
virus, which can be associated with potentially severe CNS
disease, notably in neonates and immunosuppressed patients
[35]. The most distinctive lesions of chikungunya-associated
neonatal encephalopathy were exclusively located in the
white matter and consisted of areas of reversible diffusion
restriction, a pattern classically associated with transient
ischemia with cytotoxic edema that does not imply neuronal
death [36]. These clinical and neuroradiological findings are
in agreement with our current investigations in an animal
model for CHIKV infection [25]. In this experimental model,
viral infection of the CNS is mainly detected at the meningeal
and ependymal levels rather than in the brain parenchyma,
and no viral-associated neuropathology is detected.
Even when complicated by shock or DIC syndrome with
gastrointestinal or cerebral bleeding, severe neonatal chi-
kungunya infection—whether consisting of an encephalop-
athy or mimicking DHF—was never fatal in our series. This is
a more favorable prognosis than that reported for neonatal
DHF, which still carries high lethality in NICUs [37,38].
Nevertheless, the neurological outcome of chikungu nya
encephalopathy was unpredictable and exhibited a wide
range of p ossible sequelae, ranging from mild ocular,
behavioral, or postural deficiencies to severe cerebral palsy
with extensive white matter damage mimicking a CACH/
VWM syndrome [39].
Our identification of the first vertically transmitted CHIKV
infection led to the rapid design of a prospective study aimed
at determining the epidemiology and the modaliti es of
maternal-fetal transmission of CHIKV. This study was
facilitated by the existence of a birth register, adequate
laboratory facilities, and interdisciplinary collaborations at
GHSR. However, given the magnitude of the outbreak and
our unpreparedness to deal with such an infectious disease
crisis, some aspects of vertically transmitted CHIKV infection
remain unknown, such as pregnancy status as a factor
influencing the course of CHIKV infection and the putative
preventive effect on neonatal transmission of postponing
delivery until after resolution of maternal viremia. In
addition, the limited size of our series of neonatal cases
prevented a more exhaustive description of the disease in this
age group. Last, the absence so far of long-term follow-up of
infected neonates does not allow yet the assessment of
CHIKV-associated long-term disabilities, such as learning or
cognitive impairments, apart from those already noted at 16–
24 mo follow-up.
In conclusion, maternal–fetal chikungunya transmission is
rare at the population level, exceptional in antepartum, but is
frequent in the setting of maternal viremia around term, and
is associated with severe neonatal complications. Given our
observations that maternal–fetal transmission almost invar-
iably occurs in the setting of maternal viremia concomitant
with delivery, and that C-section does not prevent CHIKV
vertical transmission, we do not recommend the systematic
performance of C-sections on infected mothers to reduce the
risk of viral transmission, but to closely monitor viremic
parturients and deliver them in maternity facilities with
adequate obstetric and neonatal care. However, our study
does not allow the conclusion that elective C-section of
infected mothers without fetal distress would not protect
from vertical transmission. In the absence of fetal distress, the
putative preventive effect on fetal transmission of postponing
delivery until resolution of maternal viremia remains to be
determined . As already shown for other maternal–fetal
transmitted viruses including hepatitis B virus and West Nile
virus, viral neutralization in exposed neonates may constitute
a useful approach to prevent neonatal infection. Our study
did confirm, at a large scale, the kinetics of transplacental
mother-transferred CHIKV-specific IgG antibodies, which
lasted longer than 6 mo for more than 50% of La Re´union
Island infants [18], as previously demonstrated for Thai
children [40]. Whether these passively transferred antibodies
exhibit a neutralizing and protective activity remains to be
determined.
Given our observation that exposed neonates are not
symptomatic at birth but become ill before day 7, we
recommend retaining them in maternity care for a week
with serial measurements of white blood cell and platelet
counts, and transferring them to the NICU as soon as they
become symptomatic, lymphopenic, or moderately thrombo-
cytopenic.
As a consequence of the recent dramatically increased
distribution of the chikungunya vector around the globe [11],
CHIKV has the potential to cause massive outbreaks in the
future. It may indeed not only re-emerge in areas where it has
already been isolated, such as the Indian subcontinent
(currently afflicted by a considerable outbreak), but also
emerge in geographical areas where it is not known to have
circulated previously (i.e., in nonimmune populations) [41]
such as the American and European continents; this
possibility is illustrated by the recent detection of CHIKV
transmission in the Emilia-Romagna region of Italy [42].
Public health agencies and clinicians should be aware of the
existence of maternal–fetal transmission of chikungunya and
be prepared to diagnose and treat this severe neonatal
infection.
Supporting Information
Alternative Language Abstract S1. Abstract Translated into French by
Patrick Ge´rardin
Found at doi:10.1371/journal.pmed.0050060.sd001 (22 KB DOC).
PLoS Medicine | www.plosmedicine.org March 2008 | Volume 5 | Issue 3 | e600421
Maternal-Fetal Chikungunya Virus Infection
Acknowledgments
We thank the medical and nursing staffs of the Department of
Obstetrics and Gynecology, and those of the Neonatal Intensive Care
Unit for their dedicated contributions to patient management and
their participation in the data collection. P. Ge´rardin thanks Pierre-
Yves Ancel for helpful advices in statistics. FAS, TC, and ML also
thank Fe´lix Rey and the Pasteur chikungunya research group for their
support.
Author contributions. P. Ge´rardin, G. Choker, and P.Y. Robillard
participated in the clinical management and the data collection for
pediatric cases. G. Barau, Y. Lenglet, Y. Touret, and A. Bouveret did
the same for parturient women. M. Bintner and S. Blanc performed
neuroimaging. H. Randrianaivo performed autopsies on fetuses. A.
Michault and K. Le Roux developed the RT-PCR and adapted the IgM
serology assays. P. Grivard helped A. Michault with virology. I.
Schuffenecker validated the RT-PCR and IgM serology assays. F.
Arenzana-Seisdedos helped set up the collaboration between the
Institut Pasteur and the GHSR. T. Couderc and M. Lecuit performed
the plac ental and animal model studies. P. Ge´rardin and P.Y.
Robillard designed the study, and reviewed the data for consistency
and errors. P. Ge´rardin, A. Michault, and M. Lecuit analyzed the data
and wrote the manuscript with the help of G. Barau and P.Y.
Robillard.
References
1. Porterfield JH (1980) Antigenic characteristics and classification of the
Tog aviridae. In: Schlesinger R, editor. The Togaviruses. New York:
Academic Press. pp. 13–46.
2. Robinson MC (1955) An epidemic of virus disease in Southern Province,
Tanganyika Territory in 1952–53. 1. Clinical features. Trans R Soc Trop
Med Hyg 49: 28–32.
3. Ross RW (1956) The Newala epidemic. III. The virus: Isolation, pathogenic
properties and relationship to the epidemic. J Hyg (London) 54: 177–191.
4. Lumsden WHR (1955) An epidemic of virus disease in Southern Province,
Tanganyika Territory in 1952–53. 11. General description and epidemiol-
ogy. Trans R Soc Trop Med Hyg 49: 33–57.
5. Hammon W, Rudnick A, Sather GE (1960) Virus as sociated with
hemorrhagic fevers of the Philippines and Thaı
¨
land. Science 131: 1102–
1103.
6. Shah KV, Gibbs CJ Jr., Banergee G (1964) Virological investigation of the
epidemic of hemorrhagic fever in Calcutta: isolation of three strains of
Chikungunya virus. Indian J Med Res 52: 676–683.
7. Mac Intosh BM, Jupp PG, Dos-Santos I (1977) Rural e pidemic of
chikungunya in South Africa with involvement of Aedes furcifer and
baboons. S Afr J Sci 73: 267–269.
8. Diallo M, Thonnon J, Traore-Lamizana M, Fontenille D (1999) Vectors of
Chikungunya virus in Senegal: current data and transmission cycles. Am J
Trop Med Hyg 60: 281–286.
9. Schuffenecker I, Iteman I, Michault A, Murri S, Franqeul L, et al. (2006)
Genome microevolution of Chikungunya viruses causing the Indian Ocean
outbreak. PLoS Med 3: e263. doi:10.1371/journal.pmed.0030263
10. Paganin F, Borgherini G, Staikowsky F, Arvin-Berod C, Poubeau P (2006)
Chikungunya a
`
l’ıˆle de la Re´union: chronique d’une e´pide´mie annonce´e.
Presse Med 35: 641–646 .
11. Reiter P, Fontenille D, Paupy C (2006) Aedes albopictus as an epidemic
vector of chikungunya virus: another emerging problem? Lancet Infect Dis
6: 463–464.
12. Borgherini G, Poub eau P, Staikowsky F, Lory M, Le Moullec N, et al. (2007)
Outbreak of chikungunya on reunion island: early clinical and laboratory
features in 157 adult patients. Clin Infect Dis 44: 1401–1407.
13. Robillard PY, Boumahni B, Ge´rardin P, Michault A, Fourmaintraux A, et al.
(2006) Transmission verticale materno-fœtale du virus chikungun ya. Presse
Med 35: 785–788.
14. Barau G, Robillard PY, Hulsey TC, Dedecker F, Laffite A, et al. (2006)
Linear association between maternal pre-pregnancy body mass index and
risk of caesarean section in term deliveries. BJOG 113: 1173–1177.
15. Staikowsky F, Le Roux K, Schuffenecker I, Laurent P, Grivard P, et al. (2007)
Retrospective survey of Chikungunya disease in Reunion Island hospital
staff. Epidemiol Infect 136: 196–206.
16. Pialoux G, Gau
¨
zere BA, Jaureguiberry S, Strobel S (2007) Chikungunya, an
epidemic arbovirosis. Lancet Infect Dis 7: 319–327.
17. Martin DA, Muth DA, Brown T, Johnson AJ, Karabatsos N, et al. (2000)
Standardization of immunoglobulin M capture enzyme-linked immuno-
sorbent assays for routine diagnosis of arboviral infections. J Clin Biol 38:
1823–1826.
18. Grivard P, Le Roux K, Laurent P, Fianu A, Perrau J, et al. (2007) Molecular
and serological diagnosi s of Chikungunya virus infection. Pathol Biol
(Paris) 55: 490–494.
19. Pastorino B, Bessaud M, Grandadam S, Murri H, Tolou JH, et al. (2005)
Development of a TaqMan RT-PCR assay witho ut RNA extraction step for
the detection and quantification of African Chikungunya viruses. J Virol
Methods 124: 65–71.
20. Laurent P, Le Roux K, Grivard P, Bertil G, Naze F, et al. (2007)
Development of a sensitive real time reverse transcriptase PCR assay with
an internal control to detect and quantify Chikungunya virus. Clin Chem
53: 1408–1414.
21. Amiel-Tison C (1976) A method for neurologic evaluation within the first
year of life. Curr Probl Pediatr 7: 1–50.
22. Touret Y, Randrianaivo H, Michault A, Schuffenecker I, Kauffmann E, et al.
(2006) Transmission materno-foetale pre´ coce du virus Chikungunya. Presse
Med 35: 1656–1658.
23. Jad hav M, Namboodripad M, Carman RH, Carey DE, Myers RM (1965)
Chikungunya disease in infants and children in Vellore: a report of clinical
and haematological features of virologically proved cases. Indian J Med Res
53: 764–776.
24. Len glet Y, Barau G, Robillard PY, Randrainaivo H, Michault A, et al. (2006)
Infection a
`
Chikungunya chez la femme enceinte et transm ission materno-
fœtale. J Gynecol Obstet BiolReprod (Paris) 35: 578–583.
25. Couderc T, Chre´ tien F, Schilte C, Disson O, Brigitte M, et al. (2008) A
mouse model for chikungunya: Young age and inefficient type-i interferon
signaling are risk factors for severe disease. PLoS Pathog 4: e29 doi:10.1371/
journal.ppat.0040029
26. Kennedy AC, Fleming J, Solomon L (1980) Chikungunya viral arthropathy:
a clinical description. J Rheumatol 7: 231–236.
27. Rehle TM (1989) Classification, distribution and importance of arboviruses.
Trop Med Parasitol 40: 391–395.
28. Ramful D, Carbonnier M, Pasquet M, Bouhma ni B, Ghazouani J, et al. (2007)
Mother-to-child transmission of Chikungunya virus infection. Pediatr
Infect Dis J 26: 811–815.
29. Fraser JR (1986) Epidemic polyarthritis and Ross River virus disease. Clin
Rheum Dis 12: 369–388.
30. Del Giudice P, Schuffenecker I, Zeller H, Grelier M, Vandenbos F, et al.
(2005) Skin manifestations of West Nile virus infection. Dermatology 211:
348–350.
31. Fad ilah SA, Sahrir S, Raymond AA, Cheong SK, Aziz JA, et al. (1999)
Quantitation of T lymphocyte subsets helps t o distinguish dengue
hemorrhagic fever from classic dengue fever during the acute febrile
stage. Southeast Asian J T rop Med Public Health 30: 710–717.
32. Sumar mo Talago W, Asrin A, Isnuhandojo B, Sahudi A (1992) Failure of
hydrocortisone to affect outcome in dengue shock syndrome. Pediatrics 69:
45–49.
33. Asher DP, Laws HF, Hayes CG (1989) The use of intravenous gamm aglo-
bulin in dengue fever, a case report. Southeast Asian J Trop Med Public
Health 20: 549–554.
34. Nim mannitiya S, Halstead SB, Cohen SN, Margiotta MR (1969) Dengue and
chikungunya virus infection in man in Thailand, 1962–1964. Am J Trop
Med Hyg 18: 954–971.
35. O’Leary DR, Kuhn S, Kniss KL, Hinckey AF, Rasmussen AS, et al. (2006)
Birth outcomes following West Nile Virus infection of pregnant women in
the United States: 2003–2004. Pediatrics 117: 537–545.
36. Ali M, Safriel Y, Sohi J, Llave A, Weathers S (2005) West Nile virus infection:
MR imaging findings in the nervous system. Am J Neuroradiol 26: 289–297.
37. Carl es G, Pfeiffer G, Talarmin A (1999) Effects of dengue fever during
pregnancy in French Guiana. Clin Infect Dis 28: 637–640.
38. Siranavin S, Nuntnarumit P, Supapannachart S, Boonkasidecha S,
Techasaensiri C, et al. (2004) Vertical dengue infection: case reports and
review. Pediatr Infect Dis J 23: 1042–1047.
39. Fogli A, Boespflug-Tanguy O (2006) The large spectrum of eIF2B-related
diseases. Biochem Soc Trans 34: 22–29.
40. Watanaveeradej V, Endy TP, Simasathien S, Kerdpanich A, Polpprasert N,
et al. (2006) Transplacental Chikungunya virus antibody kinetics, Thailand.
Emerg Infect Dis 12: 1770–1772.
41. Charrel RN, de Lamballerie X, Raoult D (2007) Chikungunya outbreaks—
the globalization of vectorborne diseases. N Engl J Med 356: 769–771.
42. Rezza G, Nicoletti L, Angelini R, Romi R, Finarelli AC, et al. (2007)
Infection with chikungunya virus in Italy: an outbreak in a temperate
region. Lancet 370: 1840–1846.
PLoS Medicine | www.plosmedicine.org March 2008 | Volume 5 | Issue 3 | e600422
Maternal-Fetal Chikungunya Virus Infection
Editors’ Summary
Background. Chikungunya virus, an emerging infectious agent that is
transmitted by day-biting mosquitoes, was first isolated from a patient in
Tanzania in the early 1950s. Since then, major outbreaks of chikungunya
fever have occurred throughout sub-Saharan Africa and in Southeast
Asia, India, and the Western Pacific, usually at intervals of about 7–8
years. The virus causes fever, rash, severe joint and muscle pains, and
sometimes arthritis (joint inflammation). These symptoms develop within
3–7 days of being bitten by an infected mosquito. Most people recover
fully within a few weeks, but joint pain can sometimes continue for years.
There is no treatment for chikungunya fever, but the symptoms can be
eased with anti-inflammatory drugs. Preventative measures include
covering arms and legs and using insecticides to avoid insect bites and
depriving the mosquitoes of their breeding sites by draining standing
water from man-made containers near human dwellings.
Why Was This Study Done? In 2005, chikungunya fever appeared for
the first time on several islands in the Indian Ocean. On La Re
´
union
Island, the disease affected 300,000 people—more than one-third of the
population—between March 2005 and December 2006. In June 2005,
clinicians identified the first case of mother-to-child chikungunya virus
transmission (vertical transmission). Public-health officials and clinicians
need to know more about how often vertical transmission occurs and its
clinical implications to help them prepare for future chikungunya fever
outbreaks. In this study, the researchers identify and characterize all the
cases of vertical chikungunya virus transmission that occurred at the
largest hospital on La Re
´
union Island during the 2005–6 outbreak.
What Did the Researchers Do and Find? The researchers enrolled all
7,504 women who gave birth at the hospital during the outbreak and
their 7,629 children into their study. They then used ‘‘ RT-PCR’’ (which
detects the genome of virus particles during an active infection) and
‘‘ IgM serology’’ (which looks for an immune response to recent infection)
to determine which women had been infected with chikungunya virus
during their pregnancy. 678 of the new mothers had been infected
sometime between conception and a week before delivery, 22 mothers
had been infected between 7 and 3 days before delivery, and 39 had
been infected 2 days either side of delivery (the ‘‘ intrapartum’’ period).
Except for three early fetal deaths that were associated wi th
chikungunya virus infections, vertical transmission was seen only in
babies born to mothers infected with the virus intrapartum. 19 of the
babies born to these women were infected with the virus—a vertical
transmission rate of nearly 50%. The women who transmitted the virus
to their offspring had more virus in their placenta than those who did
not transmit the infection. Delivery by emergency cesarean section did
not prevent transmission. All the infected babies were born healthy but
developed fever, weakness, and pain within 3–7 days. In many of them,
the number of platelets (clot-forming particles) in their blood also
dropped dramatically. Ten babies became seriously ill—nine of them
developed brain swelling; two had bleeding into their brain. Four
children had lasting disabilities at the end of the study.
What Do These Findings Mean? These findings show that mother-to-
child transmission of chikungunya virus occurs frequently when women
are infected with the virus at the time of delivery and that newborn
children infected by this route can become very ill. Although these
results do not find that cesarean section reduces infection rates, 90% of
cesarean sections involving infected infants were performed urgently,
rather than planned. The study also provides no information about
whether delaying delivery, provided that no fetal distress is observed,
until the mother’s viral load has decreased might be beneficial. More
studies are needed to provide a complete description of both the short-
term and long-term effects of chikungunya virus infection in newborn
babies, but it is clear that clinicians should monitor babies exposed to
chikungunya virus during delivery for a week after their birth. Most
importantly, clinicians and public-health officials will need to take
account of the threat that the chikungunya virus poses to newborn
children whenever and wherever it emerges.
Additional Information. Please access these Web sites via the online
version of this summary at http://dx.doi.org/10.1371/journal.pmed.
0050060.
Read the related PLoS Medicine Perspective article
The World Health Organization provides information about
chikungunya fever and a brief description of the recent chikungunya
outbreak in the Indian Ocean (in English, French, Spanish, Arabic,
Chinese, and Russian)
The US Centers for Disease Control and Prevention has a fact sheet on
chikungunya fever
The UK Health Protection Agency also provides information about
chikungunya virus, including news on recent outbreaks
The French Institut de Veille Sanitaire (Institute for Public Health
Surveillance) has a Web page on chikungunya (in French)
The Institut Pasteur has a Web page on chikungunya research (in
French and English)
PLoS Medicine | www.plosmedicine.org March 2008 | Volume 5 | Issue 3 | e600423
Maternal-Fetal Chikungunya Virus Infection
