Dengue in French Polynesia: Major Features,
Surveillance, Molecular Epidemiology and
Current Situation∗
By
Eliane Chungue, Xavier Deparis & Bernadette Murgue
Institut Territorial de Recherches Médicales Louis Malardé
P.O. Box 30, Papeete, Tahiti, French Polynesia,
Fax : 689-43 15 90, e-mail: [email protected]
Abstract
The emergence of dengue epidemics worldwide has paralleled the expansion of the mosquito
vector Aedes aegypti, along with jet air travel and increased urbanization. All the four dengue
virus serotypes (DEN-1, 2, 3 and 4) have occurred in epidemic form during the past 50 years in
French Polynesia. The first epidemic with a known serotype was due to DEN-1 which occurred in
1944 during World War II. The disease disappeared from the Eastern Pacific after a Pacific-wide
pandemic, but a series of epidemics occurred at short intervals during two decades: DEN-3 in
1964-1965, DEN-2 in 1971, DEN-1 in 1975-1976, and DEN-4 in 1979. From 1980 to 1988, the
transmission of DEN-4 continued at a very low level until the resurgence of DEN-1 and DEN-3 in
back-to-back epidemics in 1989. In 1996, DEN-2 reappeared in Tahiti and spread further into
New Caledonia, the Cook Islands, Tonga, Samoa and Fiji.
As in most Pacific countries, epidemics with only one serotype have occurred in French
Polynesia. Each time, the genetic analysis of the causative viruses showed that the current epidemic
_________________________
∗ Reproduced from Pacific Health Dialogue 1998, 5(1):154-162 with the permission of the Editor
74
Dengue Bulletin – Vol 22, 1998
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
was due to the introduction of a genotype which was different from the viruses recovered from the
past epidemics. These observations emphasize the need for an active system of clinical and
virological surveillance for the prevention and control of epidemics, together with molecular
characterization of the viruses as part of the investigation of a dengue epidemic. As of now, a new
genotype of DEN-2, different from the one involved in the 1970s, is disseminating throughout the
Pacific region.
Keywords: DF/DHF, DEN-1,2,3,4, epidemic, molecular epidemiology, French Polynesia.
Dengue Bulletin – Vol 22, 1998
75
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
Introduction
French Polynesia, situated in the South
Pacific
Ocean,
archipelagos
Gambier,
consists
(Society,
Austral
and
of
five
Tuamotu,
Marquesas
Epidemiological features
islands) comprising 120 islands, of
Epidemic pattern
which only 66 are inhabited. Most of
Dengue
the population is concentrated in the
clinically in French Polynesia since the
Society
which
nineteenth century, with documented
encompasses the Windward and the
epidemics in 1852, 1870, 1885 and
Leeward islands. Tahiti, the largest
1902(1,2,3).
island of French Polynesia in the
dengue in Tahiti of a known serotype
Windward group, contains the Capital,
occurred in 1944 as part of the
Papeete. During the past 40 years the
Pacific-wide spread of the disease
population of French Polynesia has
caused by DEN-1 during World War
undergone a dramatic increase. Since
II(4). The disease disappeared from the
1946, it has trebled in 30 years, and it
Eastern Pacific after the pandemic and,
has doubled between 1966 and 1988.
as far as is known, did not reappear
The last census (1996) recorded a
until 1964 and 1969, when DEN-3
population total of 219 521 which is
was involved(5,6). From that time on,
unequally distributed: 74% of it is
epidemics have occurred at shorter
concentrated in the Windward Islands,
intervals: DEN-2 in 1971, DEN-1 in
especially on Tahiti, of which 77% live
1975
archipelago
fever
The
and
has
been
first
DEN-4
in
known
outbreak
1979
of
were
in the urbanized Papeete. In all areas
successively
the climate is hot and tropical. A hot
exception of the DEN-2 outbreak,
rainy season from November to April
during
alternates with a cooler, drier season
cases and three deaths were observed
from May to October. The annual
on Tahiti in 1971, mildness of the
average temperature and rainfall in
disease
Tahiti is 25.7°C and 2 m, respectively.
epidemics. A recent succession of
These
are
epidemics took place after nine years
year-round
of continued transmission of DEN-4.
climatic
consistent
with
mosquito breeding.
76
conditions
the
epidemic(7,8,9).
which
severe
characterized
Back-to-back
With the
haemorrhagic
these
epidemics
past
involving
Dengue Bulletin – Vol 22, 1998
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
DEN-1 and DEN-3 occurred in 19881989. It is noteworthy that dengue
haemorrhagic
fever/dengue
shock
syndrome (DHF/DSS) occurred in the
latter epidemic (11 fatalities) while
mildness characterized the former.
These viruses were spread throughout
the Pacific region with varying degrees
of disease severity(10).
Epidemics with only one serotype
have
occurred.
serotype
Each
replaced
the
epidemic
previous
serotype that had been transmitted
during
the
inter-epidemic
period.
Simultaneous transmission of both
serotypes was only observed for a
short period of time (2-4 months)
when
the
epidemic
serotype
was
taking place. During these periods,
co-circulations were demonstrated for
DEN-2/DEN-1 in 1975, DEN-1/DEN-4
in 1979, DEN-1/DEN-3 in 1989, and
DEN-3/DEN-2 in 1996. The endemic
virus was generally swept out 7 weeks
to 4 months after the recognition of
the new epidemic. This is illustrated
by Fig.1a and Fig.1b.
Dengue Bulletin – Vol 22, 1998
77
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
Figure 1a. DEN-1 and DEN-3 epidemics (1988-1990):
Monthly distribution of dengue virus serotypes
350
Dengue 1
Dengue 3
Number of isolates
300
250
200
15 0
10 0
50
0
D
J
F
M
A
I
1988
M
J
J
A
S
O
N
D
J
F
M
I
1989
A
M
J
J
A
1990
Figure 1b. DEN-2 epidemic (1996-1997):
Monthly distribution of dengue virus serotypes
350
Number of isolate
300
Dengue 3
Dengue 2
250
200
150
100
50
0
J
F
M
A
M
1996
78
J
J
A
S
O
N
D
J
I
F
M
A
M
J
J
A
S
O
1997
Dengue Bulletin – Vol 22, 1998
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
serological surveys, the proportion of
Morbidity and mortality
Although
accurate
data
regarding
morbidity are not available, a review
of
published
literature
and
allowed
unpublished
us
to
obtain
estimates of dengue infection in the
Windward Islands during the major
epidemics which occurred between
1944 and 1997 (Table 1). Estimated
morbidity rates were calculated by
using the data obtained by serological
and/or epidemiological surveys. The
estimation of the attack rates involved
antibody prevalence in cohorts of
susceptible
populations
measurements
of
and
absenteeism
in
schools and government and business
offices. The serological attack rate was
generally around 50% as determined
immediately
after
the
epidemic
asymptomatic
infections
was
estimated to be 30%(8,11,12). During an
outbreak the mortality rate is usually
low (see Table 1). The morbidity trend
of the epidemics is also illustrated by
the
number
reported
by
of
(i)
clinical
physicians
cases
(reported
cases), (ii) cases for which a request
for laboratory confirmation is made
(suspected cases), and (iii) virologically
and/or serologically confirmed cases
(Table 2). The annual incidence rate
during
the
inter-epidemic
periods
(endemic transmission) is unknown.
One study that concerned the long
inter-epidemic period between 1980
and 1987 showed an acquisition rate
of dengue antibodies of 3% per year in
a cohort of susceptible children(13).
transmission. According to clinical and
Table 1. Morbidity and mortality of dengue during major epidemics between
1944 and 1997 in the Windward Islands (French Polynesia)
Year of
epidemic
Serotype
Population
Estimated
Windward Is.
morbidity
(1000’s)
rate(%)
No.of
No. of fatal
cases
infected
persons
(1000’s)
1944
DEN-1
29.8
62
-
26
1964-65
DEN-3
55.2
20
-
ND*
1969
DEN-3
79.1
ND*
-
ND*
Dengue Bulletin – Vol 22, 1998
79
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
1971
DEN-2
84.5
50
3
42.2
1975-76
DEN-1
94.6
25
-
34.1
1979
DEN-4
104.2
25
-
37.5
1988-89
DEN-1
140.3
17
-
35.1
1989-90
DEN-3
143.9
25
11
49.5
1996-97
DEN-2
162.6
19
1
43.5
*ND: not determined.
Table 2. Dengue cases reported during the epidemics in French Polynesia, 1964-1997
Year of epidemic
Serotype
1964-65
Number of
Reported cases*
Suspected cases
Confirmed cases
DEN-3
1358
–
–
1969
DEN-3
72
–
–
1971
DEN-2
12943
–
–
1975-76
DEN-1
2032
2018
694
1979
DEN-4
7489
1783
630
1988-89
DEN-1
6034
4836
1976
1989-90
DEN-3
6330
5583
1357
1996-97
DEN-2
7230
4424
2027
– data not available; *Institut Malarde & Direction de la Sante
Transportation and spread
of virus
In addition to the lack of effective
mosquito control, the origin of the
Pacific-wide
pandemic
of
DEN-1
(1941-1945) was very likely related to
80
intensive
movements
of
human
populations among and into areas that
were
permissive
to
dengue
transmission during World War II.
Hence, in 1944, an extensive epidemic
of dengue occurred in the Society
archipelago,
French
Polynesia.
Dengue Bulletin – Vol 22, 1998
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
Together with entomological surveys
increased
in other Pacific islands(14,15,16), the
1960s,
evidence of experimental transmission
ecology of dengue in French Polynesia
of dengue virus between monkeys
occurred after the construction of the
seaport
activity.
In
the
important
changes
in
the
suggested that Aedes polynesiensis
international airport in 1960-1961.
served as a natural vector in the past
This event suddenly brought French
French
Polynesia(14).
Polynesia into the modern era, with
the
geographical
accelerated
epidemics
in
Conversely,
urbanization
that
distribution of the disease and the
paralleled an increasing number of
entomological
dengue outbreaks. The increase of
surveys
carried
out
during the post-war epidemics (1964-
population
1996) were more consistent with the
surrounding localities comprised of
role of Ae. aegypti as a vector. For
workers recruited from the rural part
instance,
while
of
reported
in
Ae.
1953
aegypti
in
was
Tahiti
and
Papeete
from
and
its
neighbouring
its
islands. Between 1956 and 1988 the
presence had expanded progressively
population of Tahiti rose from 50% to
throughout French Polynesia, which
76% of the total population of French
was coincidental with the advent of
Polynesia; of these, 79% resided in
inter-island air connections: in the
urban areas. The expansion of Papeete
1970s in the Tuamotu, in 1982-1986
led to the growth of adjacent suburbs.
in the Marquesas Islands, and in
At this time, the metropolitan Papeete
1979-1986 in the Austral
Tahiti,
in
Islands(17).
The size and frequency of dengue
epidemics are related directly to social
lied along a 40 km coastal border only
a half kilometre wide.
The
first
re-introduction
of
and economic development, especially
dengue viruses was related to multiple
urbanization(18,19).
factors: increased jet air travel, high
first
urbanized
became
so
in
For instance, the
locale,
1850
Papeete,
the
urban areas, and lack of mosquito
European colonization. Coincidentally,
control. Papeete was hit by a DEN-3
the first occurrence of dengue on
epidemic in 1964, as was Makatea
Tahiti was reported in 1852. The
island, a nearby island of the Tuamotu
disease was limited to Papeete which
archipelago whose urban areas were
had been recently urbanized with an
subjected
Dengue Bulletin – Vol 22, 1998
during
density of susceptible population in
to
exploitation
of
81
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
phosphates.
exchanges
of the epidemics may be seen in the
existed between these two sites, in
dynamics of the recent epidemics. Fig.
both
of
Regular
Ae.
which
aegypti
were
2
shows
clearly
that
the
DEN-1
abundant. The limited transmission
epidemic in 1988-1989 moved from
within these areas was apparently
the Windward Islands to the Leeward
related to the sparse distribution of
Islands and on to the Marquesas and
Ae. aegypti in most islands of French
Austral Islands(21). Furthermore, DEN-
Polynesia(5).
A flare-up of DEN-3 was
3 was first detected in the tourist
Papeete(6).
island of Bora Bora in the Leeward
reported in 1969 in
The
speed
and
frequency
of
human exchanges by jet air travel
make possible the introduction of
dengue viruses by a dengue-infected
traveller from hyperendemic areas. A
parallel has been observed between
the spread of dengue viruses and the
route and frequency of air travel
between
the
infected
and
the
permissive areas within the Pacific
region(20). This was exemplified by the
epidemics
which
occurred
in
the
Pacific islands in the 1970s, which
always emerged from countries/areas
which
traffic.
had
The
secondarily
neighbouring
intensive
disease
from
international
has
these
island
spread
points
countries.
to
For
instance, in 1979 the first occurrence
of DEN-4 outside Asia was observed
on Tahiti. Three years later, DEN-4
had spread to many islands of the
Pacific. At the country level, in French
Polynesia, the geographical diffusion
82
Islands, and spread subsequently to
Tahiti,
then
to
archipelagoes(21).
the
These
remote
situations
were consistent with the fact that
inter-island
air
travel
was
more
intense
between
Tahiti
and
the
Leeward
Islands
than
with
the
Marquesas and Austral Islands.
Clinical features
The disease has been generally mild.
However, dengue illness caused by all
four
serotypes
accompanied
by
is
naturally
haemorrhagic
manifestations(22). In most epidemics,
a proportion of cases presented with
minor
haemorrhagic
signs:15%
in
1964 (DEN-3), 4% in 1969 (DEN-3), 3%
in 1971 (DEN-2), 15% in 1975-1976
(DEN-1), 5% in 1989 for epidemic
DEN-1, 14% for epidemic DEN-3, and
18% in 1996-1997 for DEN-2. The
two
occurrences
of
severe
Dengue Bulletin – Vol 22, 1998
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
Figure 2. DEN-1 epidemic (1988-1989): Geographical distribution
of reported cases, by week
Windward Is.
Number of reported case
500
Leeward Is.
400
Marquesas and Austral Is.
300
200
100
0
51
52
1
2
3
4
5
6
7
8
9
10
11
12
13
Calendar week 1988-1989
manifestations
hospitalization
requiring
were
in
1971
(33
cases, 3 fatalities) and 1989-1990
(401 cases, 11 fatalities). The 1971
epidemic involved mostly adults (63%)
while children represented 69% of the
cases in 1989-1990, an outbreak in
which
haemorrhagic
manifestations
were observed among 59% of the
hospitalized children. In 1996, among
the 232 hospitalized children, the
number of severe forms was low (19
cases), and the proportion of children
presenting
with
haemorrhagic
manifestations was 36%. Only one
death (adult) was reported(5,6,7,8,10,23).
Socioeconomic impacts
In islands with limited populations,
dengue fever is generally an epidemic
disease. In Asia, where the occurrence
of DHF/DSS is frequent, the social and
economic cost is high (US $31.48
million
per
year
in
Thailand(24)).
However, epidemics of the classical
disease are not to be overlooked. For
instance, during the last epidemic of
DEN-1 in French Polynesia, of the 161
subjects from whom a reply to the
question about the number of days
absented from work was obtained,
125 indicated a 3-5 days of absence.
Dengue Bulletin – Vol 22, 1998
83
14
15
16
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
Moreover, in the 1996-97 DEN-2
Commission) and local authorities, a
epidemic,
network
the
direct
costs
were
of
dengue
control
and
estimated to be US $2.5 million,
surveillance was set up. Subsequent
including
funding supported the development of
laboratory
costs,
hospitalization, and medical care. The
indirect costs, including morbidityrelated absenteeism from work, the
cost of lost production and prevention
and control activities was around US
$1.98 million. The estimated total cost
worked around US $4.48 million (i.e.
US $20
per
study,
it
inhabitant).
was
evident
From
this
that
the
estimated total cost of the 1989-90
DEN-3
DHF/DSS
epidemic,
cases,
including
was
US $40
the
per
inhabitant, whereas the estimated cost
of the 1988-89 DEN-1 epidemic was
US $15 per inhabitant. Estimation of
the costs for loss of tourism was not
made.
a
community-based
vector
control
programme. Despite the implementation
of
preventive
measures,
an
epidemic occurred a short time later.
Epidemic control involved adulticiding
of mosquitoes using ultra low volume
(ULV)
spraying
of
insecticides.
Larvicidal source reduction measures
involved the destruction or treatment
by
insecticides
of
breeding
sites.
Educational programmes accompanied
these measures(8). During the 1979
DEN-4
epidemic,
similar
measures
were implemented. After these series
of epidemics, the control programme
was maintained at a minimum level
which
comprised
continued
health
education and limited entomological
Surveillance, prevention
and control
surveillance at the international airport
and at a few sentinel stations.
Finally, the only constant and
Before 1975, the DHF epidemics did
immediately
available
not benefit from any special control
system
the
programme.
generally
system. Laboratory capabilities were
explosive and ended after exhaustion
reinforced, and modern and rapid viral
of the susceptible population. In 1975,
and serological analyses were made
Tahiti was expected to have DEN-1
available.
virus reintroduced from Fiji. With the
suspected DHF/DSS was carried out by
efforts of both regional (South Pacific
reverse
84
They
were
is
Urgent
surveillance
laboratory-based
diagnosis
transcriptase
of
polymerase
Dengue Bulletin – Vol 22, 1998
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
chain reaction (RT-PCR), allowing a
Caledonia(28). Subsequently, DEN-2 was
result within as little as one day(25).
detected instead.
Since 1988, the surveillance of dengue
fever
has
been
based
on
the
monitoring of (i) cases reported by
physicians; (ii) suspected cases; and
(iii) confirmed
cases.
Trends
of
dengue activity and the virus serotype
are continuously monitored.
The
detection
different
from
of
the
a
serotype
one
which
is
endemically transmitted constitutes a
very
important
feature
while
considering whether further epidemic
transmission
might
occur.
A
retrospective analysis of the situation
Actually, the weekly incidence of
in French Polynesia in recent years
suspected cases constitutes a valuable
shows that a new epidemic followed
Since
each detection of a new serotype in a
all the diagnoses are made only in our
sizeable susceptible population in a
laboratory, the data are immediately
country where mosquitoes were always
available. A sudden rise in laboratory
present. In 1988, the first isolation of
requests
immediate
DEN-1
in
Tahiti
sentinel
during
a
nine-year
indicator of dengue
activity(26,27).
precipitates
telephone
survey
an
among
island
intervened
inter-epidemic
physicians. In addition, through a small
period of a low-level incidence of DEN-
group of selected sentinel physicians,
4(11). The epidemic peaked a few weeks
any increase in dengue-like illness is
after,
reported
and
the
disease
spread
investigated.
Blood
throughout most of the islands of
obtained
from
French Polynesia. The first isolation of
representative cases and processed to
DEN-3 was observed in April 1989 in
detect
Bora-Bora island, then in June 1989 in
and
specimens
are
dengue
infection
by
virus
isolation or RT-PCR and/or dengue IgM
Tahiti
antibody
recognized
detection.
In
addition,
island.
The
in
epidemic
August
Furthermore,
Pacific region is taken into account in
epidemic peaked in January 1997 after
order to stimulate awareness in the
the
medical community or to set up timely
laboratory survey in August 1996.
sentinel surveillance, as happened in
Thus, virological surveillance is of
1996 when DEN-4 was detected in New
importance
case
for
recent
Tahiti(21).
information on dengue activity in the
first
the
in
was
was
an
DEN-2
detected
early
by
warning
surveillance system.
Dengue Bulletin – Vol 22, 1998
85
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
epidemiology
Molecular epidemiology
By using molecular techniques, the
transmission pathways of the viruses
of
DEN-2
viruses
is
particularly worthy of emphasis.
Nucleotide
sequencing
of
have been suggested and seem to
different parts of the genome as short
corroborate
fragments or entire genes (prM, E or
with
epidemiology.
sequence
the
descriptive
Analyse
of
relatedness
genome
NS3)
allowed
several
authors
to
demonstrated
perform comparative analysis of a
genotype groupings among all four
large variety of dengue virus strains.
DEN virus serotypes(29). Thus, one to
Sequence data obtained recently in
five genotypes were defined among
our laboratory or those retrieved from
virus
different
published databases were compared
geographic areas and over periods
within each serotype virus. Hence,
ranging from 33 to 53 years. Each
fragments of the E protein gene of (i)
genotype seems to have a defined
180 nucleotides (nt 82 to 261) from
focus of endemicity. However, certain
DEN-1 and DEN-4 viruses, (ii) 198
genotypes appeared to have been
nucleotides (nt 85 to 282) from DEN-2
transported to another part of the
viruses, and (iii) 195 nucleotides (nt
world where they became established.
73 to 267) from DEN-3 viruses were
Different
compared(29).
strains
isolated
in
transmission
pathways
of
A
divergence
of
6%
these viruses around the world are
within the studied region was taken as
pointed out, and co-circulation of two
a cut-off point for virus groupings.
or
more
genotypes
in
the
same
geographic region has been observed,
especially for DEN-2 viruses. The first
genetic evidence of the existence of a
sylvatic cycle of dengue virus, clearly
different from outbreak viruses, was
demonstrated, and further confirmed
by molecular analysis(30,31). Owing to
the
current
circulation
of
dengue
viruses in our region, the molecular
86
Three
defined
genotype
for
DEN-1
groups
viruses,
were
and
clustering of virus isolates for which
linkages
would
epidemiological
observed(32).
be
expected
grounds
Genetic
on
was
relationships
were detectable over a 50-year span
(Hawaii, 1944, to the Indian Ocean,
1993). For instance, earlier epidemic
strains from the Pacific (1974-1978)
Dengue Bulletin – Vol 22, 1998
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
clustered in genotype 1 while recent
strain derived from mutation through
epidemics
French
silent transmission. Furthermore, a
Polynesia and New Caledonia (1988-
high level of dissemination of DEN-1
1989)
strains
from
and
the
Comoro
Islands
genotype 2 during recent years, as
(1993),
as
well
Guiana,
Guadeloupe,
as
French
shown by the clustering of the viruses
Rico,
recently isolated in countries that
from
Puerto
Brazil, Peru, Nicaragua and Cuba, fell
have
in genotype 2 (as do the American
countries
strains). Therefore, it is likely that the
Pacific territories, Comoro Islands)
recent epidemics of DEN-1 in the
was
Pacific
direction
region
are
due
to
the
introduction of a new variant of virus
ties
with
French
(French
suggested,
of
tropical
Guiana,
French
although
the
spread
was
the
not
determined.
rather than the
re-emergence
a
Figure
3. Molecularof
epidemiology
of DEN-2 viruses
TORRES STRAIT 96
BURKINA FASO
1
2
83
SOMALIA
84
INDONESIA
76
SRI LANKA
SRI LANKA
SRI LANKA
85a
85b
90a
SRI LANKA
SRI LANKA
89
90b
PHILIPPINES
83
TAIWAN
87
BURMA
76
THAILAND
64
NEW GUINEA
44
CUBA
81
THAILAND
80d
JAMAICA
83
VIETNAM
87
VIETNAM
96
Dengue 2
FRENCH GUIANA 90
3
4
5
TAHITI
96
COOK
97
NEW CALEDONIA96
QUEENSLAND
92
PUERTO RICO
69
TONGA
74
INDIA
57
TAHITI
75
TRINIDAD
57
SENEGAL
74
SENEGAL
81
BURKINA FASO
80
0
2
4
6
8
10
12
14
16
18
20
% Divergence
Dengue Bulletin – Vol 22, 1998
87
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
The maximum divergence over
Samoa isolates in the same genotype.
the E protein gene fragment was
This new genotype is significantly
approximately
20%
among
DEN-2
viruses. The dendrogram shown in
Fig. 3 depicted five genotypic groups
and agreed generally with previously
published phylogenetic trees(30,31,33).
Two
involved
genotypes
in
the
have
The
Jamaican genotype (genotype 2) was
confirmed as to be related to virus
strains from South-East Asia (97.5%
similarity with Vietnamese 1974 and
1996
strains).
As
mentioned
by
Guzman et al.(33), the Cuban strain
the old New Guinea C strain (1944)
and clustered in the same genotype 2.
The Puerto Rican strain is shown to be
transmitted in the Caribbean, Central
America, India, and the South Pacific
(Tonga 1974, Tahiti 1975) within the
genotype 4. Interestingly, the recent
isolated
during
the
actual
epidemic on Tahiti in 1996 falls into a
new genotype (genotype 3) together
with
recent
isolates
from
New
Caledonia. Data sequence from D.
Phillipps
(personal
communication)
allowed also the classifycation of 1992
Queensland, 1997 Cook Islands and
88
the
previous
virus
groups (9% divergence with genotype
4). The Tahiti 1996 strain presented a
12.5%
the
difference
evidence
introduced
with
the
earlier
of
a
rather
virus
than
recently
to
the
hypothesis of a new variant derived
from earlier virus. Its closer genotype
is genotype 1, in which isolates from
Torres Strait (D. Phillipps, personal
communication), Burkina Faso or Sri
Lanka fell.
(1981) was closer (98% similarity) to
virus
from
Tahitian virus (1975), and agreed with
been
Caribbean.
distinct
Four
genotypes
were
defined
among 30 isolates of DEN-3 viruses
with the entire gene or restriction
analysis(34). The first group contained
isolates from the South Pacific (1988
to
1995),
Singapore
(1973)
and
Indonesia (1973 to 1991). The second
group comprised the viruses from
Asia (1956 to 1995) including the
reference
strain
H-87
and
the
Vietnamese strains (1974 and 1995).
The third was composed of one isolate
from
Thailand
(1971).
The
fourth
genotype included the early strains
from French Polynesia (1964 to 1969)
and
from
Puerto
Rico
(1963).
Dengue Bulletin – Vol 22, 1998
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
Maximum divergence over the studied
the three other serotypes since the
region was approximately 12% and
maximum divergence was 20% for
corresponded to the distance between
DEN-2 viruses, 12% for DEN-3 viruses,
the early (1964 and 1969) and recent
and 6.9% for DEN-1 viruses. However,
(1989
microheterogeneity
to
1995)
Polynesian/New
was
observed
Caledonian DEN-3 strains. Hence, it is
within DEN-4 geographic strains that
unlikely that the recent isolates result
clustered
from
the
Interestingly, the recent isolate from
remaining endemic virus, since no
New Caledonia (1996) appears to be
DEN-3 virus has been isolated within a
more closely related to the viruses
20-year
from
a
genetic
period.
mutation
of
Therefore,
as
for
in
two
Indonesia.
subgroups.
This
variant
was
DEN-1, these data suggest that the
present for only a few weeks in early
recent epidemics in the Pacific are due
1996.
to the introduction of a new variant
transmitted intensively is related to a
virus rather than the re-emergence of
weak adaptation of the virus to its
a
mutation
vector and /or host or to unfavourable
through silent transmission. Moreover,
climatic conditions, combined with an
there is a sequence similarity between
active vector control programme, is
a
difficult
strain
New
derived
Caledonian
from
strain
(1988)
Whether
to
recovered four months before the
unpublished
recognition
transportation
of
the
epidemic.
This
its
failure
state
data).
of
virus
to
(M.
be
Laille,
Moreover,
from
one
the
continent to another is illustrated by
emergence of DEN-3 epidemics in the
the isolation of the Haiti 1981 strain in
Pacific
Senegal from a patient who had just
favours
the
from
hypothesis
Indonesia
of
via
New
Caledonia after several months of
latency.
DEN-4 viruses seem to occur as a
single genotype around the world. The
sequence variation in the same region
of the E protein gene among DEN-4
viruses was lower (4.9%) than among
Dengue Bulletin – Vol 22, 1998
arrived from Haiti(31).
These studies demonstrated that
dengue viruses could be identified and
classified in genotypic groups that
may circulate concurrently in the same
region. Moreover, in some instances,
the origin of the newly introduced
89
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
virus
has
been
suggested.
For
instance, close relationships between
the
Polynesian
strains
(1964
strain
or
between
is
involved
in
the
pathogenicity of DHF/DSS.
and
1969) and the Puerto Rican (1963)
DEN-3
change
the
Current situation
Polynesian (1975) and Puerto Rican
Fig. 4 illustrates the yearly distribution
(1969) strains or Trinidad (1957) DEN-
of
2
virus
dengue cases from 1988 to 1997.
exchanges between the Caribbean and
Since its recent introduction, DEN-2
the South Pacific owing to the frequent
has been continuously transmitted. In
trades from Europe via the Panama
view of the risk of epidemic dengue,
channel. The South Pacific-American
when considering the time elapsed
connection was also suggested by the
since the last epidemic of a given
close genetic relationships between
serotype
DEN-4 viruses (99 to 100% homology).
population born subsequently, French
In more recent years, the introduction
Polynesia is at high risk for the
of a new virus in the South Pacific
reintroduction of DEN-4, and to a
seems to be more related to frequent
lesser extent, of DEN-1. Retrospective
air travel exchanges with Indonesia
observations
(DEN-3 in 1989, and DEN-4 in 1996).
epidemiology
Each time, a new epidemic was due to
recent
the introduction of a new genotype.
emergency adulticiding and larvicidal
The current DEN-2 virus spreading
control
from Tahiti to New Caledonia and the
immediately after the detection of the
Cook Islands belongs to the same
first
genotype.
the
Advantage may be taken of the usual
viruses
one-to- several months of the time
strains,
efficient
suggest
This
possible
emphasizes
dissemination
of
the
intra-regional
epidemic.
Furthermore,
been associated with severe disease
potential. However, it is still unclear
whether
90
any
particular
molecular
the
have
before
In
confirmed
size
of
the
concerning
and
control
the
the
of
the
showed
to
emergence
lag
travel.
and
and
epidemics
among these island countries through
certain genotypes of the viruses have
suspected
of
be
that
applied
either
recognition
conclusion,
virus.
of
an
improved
laboratory-based surveillance, genetic
investigation
disease
of
circulating
surveillance
viruses,
(specific
and
Dengue Bulletin – Vol 22, 1998
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
febrile
diseases),
programmes,
health
and
education
vector
D.
control
Pons
for
their
secretarial
assistance.
programmes must be promoted for the
better
prevention
and
control
of
References
Figure 4. Yearly distribution of dengue in French Polynesia,
1989-1997
60%
9000
Suspected cases
Confirmed cases
% confirmed cases
Number of cases
8000
7000
50%
40%
6000
5000
30%
4000
20%
3000
2000
10 %
10 0 0
0
0%
19 8 8
19 8 9
19 9 0
19 9 1
19 9 2
epidemics.
19 9 5
19 9 6
19 9 7
DEN-2
1. Chassaniol DMP. (1885). Lettre de Mr. Le
Chef de Service de la Santé à Mr. Le
Acknowledgments
authors
19 9 4
DEN-3
DEN-1
The
19 9 3
thank
C.
Directeur de l'Intérieur. Bull. Soc. Etudes
Roche,
O.
Cassar, F. Laur, A. Babinet, K. Laille
and JF Schnee for their collaboration,
and D.Q. Ha of Pasteur Institute, Ho
Chi Minh City and M. Laille from
Institut Pasteur de Nouvelle Calédonie
for providing the 1996 virus strains.
The authors also thank N. Maruhi and
Dengue Bulletin – Vol 22, 1998
Océaniennes-EFO,
juin
1885;
Numéro
spécial, 231, tome XIX (8): 218.
2. Buisson. Les Marquises et les marquisiens.
Ann Hyg Colon, 1903; 6: 535-551. Hodée P.
In : Archevéché de Papeete (Ed) Tahiti 18341984, 150 ans de vie chrétienne en église.
Saint-Paul, Paris/Fribourg 1983.
3. Rosen L. Dengue antibodies in residents of
the Society Islands, French Oceania. Am J
Trop Med Hyg, 1958; 7: 403-405.
91
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
4. Laigret J, Rosen L, Scholammer G. Sur une
épidémie de Dengue survenue à Tahiti en
1964.
Relations
avec
les
"Fièvres
Hémorragiques" du Sud-Est Asiatique. Bull
Soc Path Exot, 1967; 60: 339-353.
5. Saugrain J, Rosen L, Outin-Fabre D, Moreau JP. Une récente épidémie d'arbovirose du type
Dengue
à
Tahiti.
Comparaison
avec
l'épidémie de 1964. Bull Soc Path Exot, 1970;
63: 631-41.
6. Moreau J-P, Rosen L, Saugrain J, Lagraulet J.
An epidemic of Dengue on Tahiti associated
with hemorrhagic manifestations. Am J Trop
Med Hyg, 1973; 22: 237-241.
7. Kaueffer H, Pichon G, Merlin M, et al. A
of
immunoglobulin
immunosorbent
G
assay
enzyme-linked
(IgG-ELISA)
and
haemagglutination inhibition (HI) test for the
detection of Dengue antibodies. Prevalence of
Dengue IgG-ELISA antibodies in Tahiti. Trans
Roy Soc Trop Med Hyg, 1989; 83: 708-711.
13. Rosen L, Rozeboom LE, Sweet BH, Sabin AB.
The
transmission
of
Dengue
by
Aedes
polynesiensis marks. Am J Trop Med Hyg,
1954; 3: 878-882.
14. Hargrave WW. Report of Dengue epidemic in
American Samoa. US Nav Med Bull, 1931; 29:
565-572.
15. Mumford
EP,
Mohr
JL.
Manual
of
the
distribution of communicable diseases and
propos d'une épidémie contrôlée de Dengue
their vectors in the tropics. Pacific Islands
en Polynésie Française. Méd Trop, 1976; 36:
Section-Part I, 26 pp ; Suppl. to Am J Trop
455-459.
8. Parc F, Tetaria C, Pichon G, et al. La Dengue
due au virus de type 4 en Polynésie Française.
II. – Observations biologiques préliminaires
sur quelques points précis d’épidémiologie et
de physiopathologie. Méd Trop, 1981; 41:
97-102.
9. Chungue E, Laudon F, Glaziou P. Dengue and
Dengue
haemorrhagic
fever
in
French
Polynesia - current situation. Trop Med,
1993; 35: 209-215.
10. Chungue E, Burucoa C, Boutin J-P. Dengue 1
epidemic in French Polynesia, 1988-1989:
surveillance
and
clinical,
epidemiological,
virological and serological findings in 1752
documented clinical cases. Trans Roy Soc Trop
Med Hyg, 1992; 86: 193-197.
11. Chungue E, Glaziou P, Spiegel A. Estimation
of Dengue infection attack rate in a cohort of
children during a DEN-3 outbreak in Tahiti.
Southeast Asean J Trop Med Public Health,
1992; 23: 157-158.
92
12. Chungue E, Marché G, Plichart R. Comparison
Med Hyg, 1944; 24.
16. Séchan
Y,
Lardeux
F,
Loncke
S.
Les
arthropodes vecteurs de maladies agents de
nuisances. In : ORSTOM (ed.) Atlas de
Polynésie Française, 1993, Planche 58.
17. Fages J, Pichon G. Dengue and urbanization
in
French
Polynesia.
Pacific
Science
Association, Honolulu, Hawaii, 1975; 281286.
18. Gubler DJ. Aedes aegypti and Aedes aegyptiborne disease control in the 1990s: top down
or bottom up. Am J Trop Med Hyg, 1989; 40:
571-578.
19. Lyon M-F. Epidémiologie de la Dengue dans
la région du Pacifique Sud. Thèse, Doctorat
en Médecine, Faculté de Médecine, Paris
1980.
20. Chungue E, Spiegel A, Roux J. DEN-3 in
French Polynesia: preliminary data. J Med
Aust, 1990; 152: 557-558.
21. Halstead SB. Antibody, macrophages, dengue
virus infection, shock, and hemorrhage: a
Dengue Bulletin – Vol 22, 1998
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
pathogenetic cascade. Rev Infect Dis, 1989;
11, supp. 4: S830-S839.
22. Glaziou P, Chungue E, Gestas P. Dengue fever
and dengue shock syndrome in French
Polynesia. Southeast Asian J Trop Med Public
Health, 1992; 23: 531-532.
23. Knudsen
AB.
Global
strategy
for
the
prevention and control of dengue and dengue
haemorrhagic fever. In : Saluzzo JF, Dodet B
(eds) Factors in the emergence of arbovirus
diseases. Elsevier, 1997; pp 131-140.
épidémiologique de la dengue en 1996.
Bises, 1997; 2: 5-8.
28. Chungue,
E.
Molecular
epidemiology
of
dengue viruses. In : Saluzzo JF, Dodet B (eds)
Factors
in
the
Emergence
of
arbovirus
diseases. Elsevier, 1997; pp 93-101.
29. Rico-Hesse R. Molecular evolution of dengue
serotype 1 and 2 in nature. Virol, 1990; 174:
479-493.
30. Deubel V, Nogueira R, Drouet MT. Direct
sequencing of genomic cDNA fragments
24. Chungue E, Roche C, Lefevre M-F, Barbazan
amplified by the polymerase chain reaction
P, Chanteau S. Ultra-rapid, simple, sensitive,
for molecular epidemiology of Dengue-2
of dengue viral RNA from clinical specimens
31. Chungue E, Cassar O, Drouet MT. Molecular
and economical silica method for extraction
and mosquitoes by reverse transcriptase-
polymerase chain reaction. J Med Virol, 1993;
40: 142-145.
25. Chan KL, Ng SK, Chew LM. The 1973 dengue
haemorrhagic fever outbreak in Singapore
and its control. Sing Med J, 1977; 18: 81-93.
viruses. Arch Virol, 1993; 129: 197-210.
epidemiology of Dengue-1 and Dengue-4
viruses. J Gen Virol, 1995; 76: 1877-1884.
32. Guzman MG, Deubel V, Pelegrino JL. Partial
nucleotide and amino acid sequences of the
enveloppe/nonstructural
protein-1
gene
junction of four Dengue-2 virus strains
26. Chungue E, Boutin J-P, Roux J. Dengue
isolated during the 1981 Cuban epidemic.
to use the excess number of laboratory
33. Chungue E, Deubel V, Cassar O. Molecular
surveillance in French Polynesia: an attempt
requests
for
confirmation
of
Dengue
diagnosis as an indicator of dengue activity.
Eur J Epidemiol, 1991; 7: 616-620.
27. Deparis X, Chungue E, Murgue B. Bilan du
réseau
sentinelle
Dengue Bulletin – Vol 22, 1998
de
surveillance
Am J Trop Med Hyg, 1995; 52: 241-246.
epidemiology of DEN-3 viruses and genetic
relatedness among DEN-3 strains isolated
from patients with mild or severe form of
Dengue fever in French Polynesia. J Gen Virol,
1993; 74: 2765-2.
93