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 Fax : 689-43 15 90, e-mail: echungue@malarde.pf P.O. Box 30, Papeete, Tahiti, French Polynesia,
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 Gambier, Ocean, Austral consists (Society, and of five archipelagos Tuamotu, Marquesas
Epidemiological features Epidemic pattern Dengue fever has been known clinically in French Polynesia since the nineteenth century, with documented epidemics in 1852, 1870, 1885 and 1902(1,2,3). The first outbreak of dengue in Tahiti of a known serotype occurred in 1944 as part of the Pacific-wide spread of the disease caused by DEN-1 during World War II(4). The disease disappeared from the Eastern Pacific after the pandemic and, as far as is known, did not reappear until 1964 and 1969, when DEN-3 was involved(5,6). From that time on, epidemics have occurred at shorter intervals: DEN-2 in 1971, DEN-1 in 1975 and DEN-4 in 1979 were successively during epidemic(7,8,9). severe With the
islands) comprising 120 islands, of which only 66 are inhabited. Most of the population is concentrated in the Society archipelago which encompasses the Windward and the Leeward islands. Tahiti, the largest island of French Polynesia in the Windward group, contains the Capital, Papeete. During the past 40 years the population of French Polynesia has undergone a dramatic increase. Since 1946, it has trebled in 30 years, and it has doubled between 1966 and 1988. The last census (1996) recorded a population total of 219 521 which is unequally distributed: 74% of it is concentrated in the Windward Islands, especially on Tahiti, of which 77% live in the urbanized Papeete. In all areas the climate is hot and tropical. A hot rainy season from November to April alternates with a cooler, drier season from May to October. The annual average temperature and rainfall in Tahiti is 25.7°C and 2 m, respectively. These climatic with conditions the are consistent year-round
exception of the DEN-2 outbreak, which haemorrhagic cases and three deaths were observed on Tahiti in 1971, mildness of the disease characterized these past epidemics. A recent succession of epidemics took place after nine years of continued transmission of DEN-4. Back-to-back epidemics involving
mosquito breeding.
76
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. replaced the Each the epidemic previous period. serotype during
serotype that had been transmitted inter-epidemic 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
300
Dengue 1 Dengue 3
Number of isolates
250
200
15 0
10 0
50
0
D 1988
J
F
M
A
M
J
J
A
S
O
N
D
J
F
M
A
M
J
J
A
I
1989
I
1990
Monthly distribution of dengue virus serotypes 350 300
Figure 1b. DEN-2 epidemic (1996-1997):
Number of isolate
Dengue 3 Dengue 2
250 200 150 100 50 0
J
F
M
A
M
J
J
A
S
O
N
D
J
F
M
A
M
J
J
A
S
O
1996
I
1997
78
Dengue Bulletin – Vol 22, 1998
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
Morbidity and mortality Although of accurate data regarding morbidity are not available, a review published and us unpublished to obtain literature allowed
serological surveys, the proportion of 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 by of (i) clinical cases reported physicians (reported
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 measurements populations of absenteeism and in
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).
schools and government and business offices. The serological attack rate was generally around 50% as determined immediately after the epidemic 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 Population Serotype Windward Is. (1000’s) DEN-1 DEN-3 DEN-3 29.8 55.2 79.1 Estimated morbidity rate(%) 62 20 ND* No.of No. of fatal cases infected persons (1000’s) 26 ND* ND*
1944 1964-65 1969
Dengue Bulletin – Vol 22, 1998
79
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
1971 1975-76 1979 1988-89 1989-90 1996-97 *ND: not determined.
DEN-2 DEN-1 DEN-4 DEN-1 DEN-3 DEN-2
84.5 94.6 104.2 140.3 143.9 162.6
50 25 25 17 25 19
3 11 1
42.2 34.1 37.5 35.1 49.5 43.5
Table 2. Dengue cases reported during the epidemics in French Polynesia, 1964-1997 Year of epidemic 1964-65 1969 1971 1975-76 1979 1988-89 1989-90 1996-97 Serotype DEN-3 DEN-3 DEN-2 DEN-1 DEN-4 DEN-1 DEN-3 DEN-2 Number of Reported cases* 1358 72 12943 2032 7489 6034 6330 7230 Suspected cases – – – 2018 1783 4836 5583 4424 Confirmed cases – – – 694 630 1976 1357 2027
– data not available; *Institut Malarde & Direction de la Sante intensive were movements permissive to of human dengue
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
populations among and into areas that transmission during World War II. Hence, in 1944, an extensive epidemic of dengue occurred in the Society archipelago, French Polynesia.
80
Dengue Bulletin – Vol 22, 1998
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
Together with entomological surveys in other Pacific islands(14,15,16), the evidence of experimental transmission of dengue virus between monkeys suggested that Aedes polynesiensis served as a natural vector in the past epidemics Conversely, entomological in French the surveys Polynesia(14). geographical carried out
increased 1960s,
seaport
activity. changes
In in
the the
important
ecology of dengue in French Polynesia occurred after the construction of the international airport in 1960-1961. This event suddenly brought French Polynesia into the modern era, with accelerated urbanization that paralleled an increasing number of dengue outbreaks. The increase of population in Papeete and its surrounding localities comprised of workers recruited from the rural part of Tahiti and from neighbouring islands. Between 1956 and 1988 the population of Tahiti rose from 50% to 76% of the total population of French Polynesia; of these, 79% resided in urban areas. The expansion of Papeete led to the growth of adjacent suburbs. At this time, the metropolitan Papeete lied along a 40 km coastal border only a half kilometre wide. The first re-introduction of dengue viruses was related to multiple factors: increased jet air travel, high density of susceptible population in urban areas, and lack of mosquito control. Papeete was hit by a DEN-3 epidemic in 1964, as was Makatea island, a nearby island of the Tuamotu archipelago whose urban areas were subjected to exploitation of 81
distribution of the disease and the during the post-war epidemics (19641996) were more consistent with the role of Ae. aegypti as a vector. For instance, reported while in
Ae.
aegypti
was its
1953
in
Tahiti,
presence had expanded progressively throughout French Polynesia, which was coincidental with the advent of inter-island air connections: in the 1970s in the Tuamotu, in 1982-1986 in the Marquesas Islands, and in 1979-1986 in the Austral Islands(17). The size and frequency of dengue epidemics are related directly to social and economic development, especially urbanization(18,19). first urbanized so in became For instance, the locale, 1850 Papeete, the during
European colonization. Coincidentally, the first occurrence of dengue on Tahiti was reported in 1852. The disease was limited to Papeete which had been recently urbanized with an Dengue Bulletin – Vol 22, 1998
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
phosphates. both of
Regular
exchanges
of the epidemics may be seen in the dynamics of the recent epidemics. Fig. 2 shows clearly that the DEN-1 epidemic in 1988-1989 moved from the Windward Islands to the Leeward Islands and on to the Marquesas and Austral Islands(21). Furthermore, DEN3 was first detected in the tourist island of Bora Bora in the Leeward Islands, and spread subsequently to Tahiti, then to the remote situations was and with more the the archipelagoes(21). inter-island intense Leeward air These travel Tahiti than
existed between these two sites, in which
Ae.
aegypti
were
abundant. The limited transmission within these areas was apparently related to the sparse distribution of
Ae. aegypti in most islands of French Polynesia(5). The A flare-up of DEN-3 was Papeete(6). frequency of and reported in 1969 in speed
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 intensive disease from island these international has spread to For points
were consistent with the fact that between Islands
Marquesas and Austral Islands.
Clinical features The disease has been generally mild. However, dengue illness caused by all four serotypes by is naturally accompanied 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
secondarily neighbouring
countries.
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
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 500 Windward Is. Leeward Is. 400 Marquesas and Austral Is. 300 200 100
Number of reported case
0 51 52 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16
Calendar week 1988-1989
manifestations hospitalization were in
requiring 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
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
Moreover, in the 1996-97 DEN-2 epidemic, including the direct laboratory costs were costs, estimated to be US $2.5 million, 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 study, DEN-3 DHF/DSS per it inhabitant). was evident From that this the the per
Commission) and local authorities, a network of dengue control and surveillance was set up. Subsequent funding supported the development of a community-based of preventive vector control an programme. Despite the implementation measures, 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 at the international airport and at a few sentinel stations. Finally, the only constant and immediately system is available the surveillance laboratory-based
estimated total cost of the 1989-90 epidemic, cases, including was US $40
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.
Surveillance, prevention and control Before 1975, the DHF epidemics did not benefit from any special control programme. They were generally explosive and ended after exhaustion of the susceptible population. In 1975, Tahiti was expected to have DEN-1 virus reintroduced from Fiji. With the efforts of both regional (South Pacific 84
system. Laboratory capabilities were reinforced, and modern and rapid viral and serological analyses were made available. reverse Urgent diagnosis of suspected DHF/DSS was carried out by transcriptase polymerase
Dengue Bulletin – Vol 22, 1998
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
chain reaction (RT-PCR), allowing a result within as little as one day(25). 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. Actually, the weekly incidence of suspected cases constitutes a valuable indicator of dengue activity(26,27). Since all the diagnoses are made only in our laboratory, the data are immediately available. A sudden rise in laboratory requests telephone precipitates survey an immediate sentinel among
Caledonia(28). Subsequently, DEN-2 was detected instead. The different very detection from the of a serotype which is one
endemically transmitted constitutes a important feature occur. while A considering whether further epidemic transmission might retrospective analysis of the situation in French Polynesia in recent years shows that a new epidemic followed each detection of a new serotype in a sizeable susceptible population in a country where mosquitoes were always present. In 1988, the first isolation of DEN-1 during in a Tahiti island intervened nine-year inter-epidemic
physicians. In addition, through a small group of selected sentinel physicians, any increase in dengue-like illness is reported specimens detect and are investigated. obtained infection In by Blood from virus
period of a low-level incidence of DEN4(11). The epidemic peaked a few weeks after, and the disease spread throughout most of the islands of French Polynesia. The first isolation of DEN-3 was observed in April 1989 in Bora-Bora island, then in June 1989 in Tahiti island. in The the epidemic in recent was recognized Furthermore, the first August Tahiti(21). DEN-2 by
representative cases and processed to dengue isolation or RT-PCR and/or dengue IgM antibody detection. addition, information on dengue activity in the Pacific region is taken into account in order to stimulate awareness in the medical community or to set up timely sentinel surveillance, as happened in 1996 when DEN-4 was detected in New
epidemic peaked in January 1997 after case was detected laboratory survey in August 1996. Thus, virological surveillance is of importance for an early warning surveillance system.
Dengue Bulletin – Vol 22, 1998
85
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
Molecular epidemiology By using molecular techniques, the transmission pathways of the viruses have been suggested and seem to corroborate epidemiology. sequence with the descriptive of genome demonstrated Analyse
epidemiology
of
DEN-2
viruses
is
particularly worthy of emphasis. Nucleotide sequencing of
different parts of the genome as short fragments or entire genes (prM, E or NS3) allowed several authors to perform comparative analysis of a large variety of dengue virus strains. Sequence data obtained recently in our laboratory or those retrieved from published databases were compared within each serotype virus. Hence, fragments of the E protein gene of (i) 180 nucleotides (nt 82 to 261) from DEN-1 and DEN-4 viruses, (ii) 198 nucleotides (nt 85 to 282) from DEN-2 viruses, and (iii) 195 nucleotides (nt 73 to 267) from DEN-3 viruses were compared(29). A divergence of 6% within the studied region was taken as a cut-off point for virus groupings. Three defined linkages genotype DEN-1 be groups viruses, expected were and on was
relatedness
genotype groupings among all four DEN virus serotypes(29). Thus, one to five genotypes were defined among virus strains isolated in different geographic areas and over periods ranging from 33 to 53 years. Each genotype seems to have a defined focus of endemicity. However, certain genotypes appeared to have been transported to another part of the world where they became established. Different transmission pathways of these viruses around the world are pointed out, and co-circulation of two 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
for
clustering of virus isolates for which would epidemiological observed(32). grounds
Genetic
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)
86
Dengue Bulletin – Vol 22, 1998
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
clustered in genotype 1 while recent epidemics 1989) (1993), Guiana, and as strains the well from Comoro as from Puerto French Islands French Rico, Polynesia and New Caledonia (1988-
strain derived from mutation through silent transmission. Furthermore, a high level of dissemination of DEN-1 genotype 2 during recent years, as shown by the clustering of the viruses recently isolated in countries that have ties with (French French Guiana, although spread was tropical French the not countries was
Guadeloupe,
Brazil, Peru, Nicaragua and Cuba, fell in genotype 2 (as do the American strains). Therefore, it is likely that the recent epidemics of DEN-1 in the Pacific region are due to the introduction of a new variant of virus
Pacific territories, Comoro Islands) suggested, of the direction
determined.
rather than the re-emergence a Figure 3. Molecularof epidemiology of DEN-2 viruses TORRES STRAIT 96 BURKINA FASO SOMALIA INDONESIA SRI LANKA SRI LANKA SRI LANKA SRI LANKA SRI LANKA PHILIPPINES TAIWAN BURMA THAILAND NEW GUINEA CUBA THAILAND JAMAICA VIETNAM VIETNAM 83 84 76 85a 85b 90a 89 90b 83 87 76 64 44 81 80d 83 87 96
1
Dengue 2
2
FRENCH GUIANA 90
3
TAHITI 96 COOK 97 NEW CALEDONIA96 QUEENSLAND 92 PUERTO RICO TONGA 69 74 57 75 57 74 81 80 0 2 4 6 8 10 12 14 16 18 20
4
INDIA TAHITI TRINIDAD SENEGAL
5
SENEGAL BURKINA FASO
% Divergence Dengue Bulletin – Vol 22, 1998 87
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
The maximum divergence over the E protein gene fragment was 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 been The
Samoa isolates in the same genotype. This new genotype is significantly distinct from the previous virus groups (9% divergence with genotype 4). The Tahiti 1996 strain presented a 12.5% the difference of with a the earlier recently to the Tahitian virus (1975), and agreed with evidence virus than introduced rather
Caribbean.
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 (1981) was closer (98% similarity) to 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 virus 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
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. 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).
88
Dengue Bulletin – Vol 22, 1998
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
Maximum divergence over the studied region was approximately 12% and corresponded to the distance between the early (1964 and 1969) and recent (1989 to 1995) Polynesian/New Caledonian DEN-3 strains. Hence, it is unlikely that the recent isolates result from a genetic mutation of the remaining endemic virus, since no DEN-3 virus has been isolated within a 20-year period. Therefore, as for DEN-1, these data suggest that the recent epidemics in the Pacific are due to the introduction of a new variant virus rather than the re-emergence of a strain derived from mutation through silent transmission. Moreover, there is a sequence similarity between a New Caledonian of the strain (1988) This the New recovered four months before the recognition favours Pacific 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
the three other serotypes since the maximum divergence was 20% for DEN-2 viruses, 12% for DEN-3 viruses, and 6.9% for DEN-1 viruses. However, microheterogeneity clustered in was two observed subgroups. within DEN-4 geographic strains that Interestingly, the recent isolate from New Caledonia (1996) appears to be more closely related to the viruses from 1996. Indonesia. Whether This its variant to was be present for only a few weeks in early failure transmitted intensively is related to a weak adaptation of the virus to its vector and /or host or to unfavourable climatic conditions, combined with an active vector control programme, is difficult to state data). of virus (M. from Laille, one unpublished transportation Moreover,
epidemic. of via
the from
hypothesis Indonesia
continent to another is illustrated by the isolation of the Haiti 1981 strain in Senegal from a patient who had just 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
emergence of DEN-3 epidemics in the Caledonia after several months of
89
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
virus the
has
been
suggested. (1964
For and the
change
is
involved
in
the
instance, close relationships between Polynesian strain strains or 1969) and the Puerto Rican (1963) DEN-3 between Polynesian (1975) and Puerto Rican (1969) strains or Trinidad (1957) DEN2 strains, suggest possible virus exchanges between the Caribbean and the South Pacific owing to the frequent trades from Europe via the Panama channel. The South Pacific-American connection was also suggested by the close genetic relationships between DEN-4 viruses (99 to 100% homology). In more recent years, the introduction of a new virus in the South Pacific seems to be more related to frequent air travel exchanges with Indonesia (DEN-3 in 1989, and DEN-4 in 1996). Each time, a new epidemic was due to the introduction of a new genotype. The current DEN-2 virus spreading from Tahiti to New Caledonia and the Cook Islands belongs to the same genotype. efficient This emphasizes of the dissemination travel. viruses
pathogenicity of DHF/DSS.
Current situation Fig. 4 illustrates the yearly distribution of the suspected and confirmed dengue cases from 1988 to 1997. Since its recent introduction, DEN-2 has been continuously transmitted. In view of the risk of epidemic dengue, when considering the time elapsed since the last epidemic of a given serotype and the size of the population born subsequently, French Polynesia is at high risk for the reintroduction of DEN-4, and to a lesser extent, of DEN-1. Retrospective observations epidemiology recent control first concerning and control showed be either of the the that applied virus.
epidemics have to of
emergency adulticiding and larvicidal immediately after the detection of the emergence Advantage may be taken of the usual one-to- several months of the time lag before the recognition of an epidemic. In conclusion, of circulating improved viruses, and
among these island countries through intra-regional Furthermore, certain genotypes of the viruses have been associated with severe disease potential. However, it is still unclear whether any particular molecular
laboratory-based surveillance, genetic investigation disease surveillance (specific
90
Dengue Bulletin – Vol 22, 1998
Dengue in French Polynesia: Major Features, Surveillance, Molecular Epidemiology and Current Situation
febrile
diseases), and
health vector and
education control of
D.
Pons
for
their
secretarial
programmes, better
assistance.
programmes must be promoted for the prevention control
References
Figure 4. Yearly distribution of dengue in French Polynesia, 1989-1997 9000 8000 60%
Number of cases
7000 6000 5000
S uspected cases Confirmed cases % confirmed cases
50%
40%
30% 4000 3000 2000 10 % 10 0 0 0 19 8 8 19 8 9 19 9 0 19 9 1 19 9 2 19 9 3 19 9 4 19 9 5 19 9 6 19 9 7 0% 20%
DEN-1
DEN-3
DEN-2
epidemics.
1. Chassaniol DMP. (1885). Lettre de Mr. Le Chef de Service de la Santé à Mr. Le Directeur de l'Intérieur. Bull. Soc. Etudes
Acknowledgments The authors thank C. 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 propos d'une épidémie contrôlée de Dengue en Polynésie Française. Méd Trop, 1976; 36: 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 12. Chungue E, Marché G, Plichart R. Comparison of immunoglobulin G enzyme-linked and immunosorbent assay (IgG-ELISA)
haemagglutination inhibition (HI) test for the detection of Dengue antibodies. Prevalence of Roy Soc Trop Med Hyg, 1989; 83: 708-711. The transmission of Dengue by Dengue IgG-ELISA antibodies in Tahiti. Trans 13. Rosen L, Rozeboom LE, Sweet BH, Sabin AB. 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 their vectors in the tropics. Pacific Islands Section-Part I, 26 pp ; Suppl. to Am J Trop Med Hyg, 1944; 24.
16. Séchan
arthropodes vecteurs de maladies agents de nuisances. In : ORSTOM (ed.) Atlas de Polynésie Française, 1993, Planche 58. in French Polynesia. Pacific
Y,
Lardeux
F,
Loncke
S.
Les
17. Fages J, Pichon G. Dengue and urbanization Association, Honolulu, Hawaii, 1975; 281286. 18. Gubler DJ. Aedes aegypti and Aedes aegyptiborne disease control in the 1990s: top down 571-578. 19. Lyon M-F. Epidémiologie de la Dengue dans en Médecine, Faculté de Médecine, Paris or bottom up. Am J Trop Med Hyg, 1989; 40:
Science
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, 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. epidemiological,
la région du Pacifique Sud. Thèse, Doctorat
1980.
20. Chungue E, Spiegel A, Roux J. DEN-3 in Aust, 1990; 152: 557-558.
French Polynesia: preliminary data. J Med
Southeast Asean J Trop Med Public Health, 1992; 23: 157-158.
21. Halstead SB. Antibody, macrophages, dengue virus infection, shock, and hemorrhage: a
92
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. AB. Global 23. Knudsen strategy for épidémiologique de la dengue en 1996.
Bises, 1997; 2: 5-8. E.
28. Chungue, Factors
dengue viruses. In : Saluzzo JF, Dodet B (eds) in the Emergence of diseases. Elsevier, 1997; pp 93-101.
Molecular
epidemiology
of
arbovirus
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.
the
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 amplified by the polymerase chain reaction for molecular epidemiology of Dengue-2 viruses. Arch Virol, 1993; 129: 197-210.
24. Chungue E, Roche C, Lefevre M-F, Barbazan P, Chanteau S. Ultra-rapid, simple, sensitive, and economical silica method for extraction of dengue viral RNA from clinical specimens and mosquitoes by reverse transcriptase40: 142-145. polymerase chain reaction. J Med Virol, 1993; 25. Chan KL, Ng SK, Chew LM. The 1973 dengue and its control. Sing Med J, 1977; 18: 81-93.
31. Chungue E, Cassar O, Drouet MT. Molecular 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 junction of four Dengue-2 virus strains isolated during the 1981 Cuban epidemic. Am J Trop Med Hyg, 1995; 52: 241-246. 33. Chungue E, Deubel V, Cassar O. Molecular 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. gene
haemorrhagic fever outbreak in Singapore
26. Chungue E, Boutin J-P, Roux J. Dengue surveillance in French Polynesia: an attempt requests for confirmation of to use the excess number of laboratory 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 de surveillance
epidemiology of DEN-3 viruses and genetic
Dengue Bulletin – Vol 22, 1998
93