280
* Address correspondence to Karin L. Schiøler, Department of International Health, Immunology and Microbiology, University of Copenhagen, CSS, Øster Farimagsgade 5, DK-1014 Copenhagen K, Denmark. E-mail: ksch@sund.ku.dkDengue Transmission in the Small-Island Setting: Investigations from
the Caribbean Island of Grenada
Karin L. Schiøler * and Calum N. Macpherson
Department of International Health, Immunology, and Microbiology, University of Copenhagen, Denmark; Windward Islands Research and Education Foundation, St. George’s University, Grenada
Abstract. The Caribbean region has experienced a major surge in dengue activity in recent decades. Yet, for many, and especially the smaller islands, the true extent and general epidemiology of dengue transmission remains unclear because of inadequate systems of surveillance and reporting. We established an active surveillance system supported by labora-tory-based diagnosis on the small island of Grenada, from January 2001 through June 2002. The surveillance data provided evidence of three distinct modes of disease dynamics in the form of low endemic, pre-epidemic, and overt epidemic trans-mission of mostly benign dengue fever. A shift in serotype activity and modal age was evident during the noted transition, with the more densely populated south end of the island presenting the focus of transmission.
INTRODUCTION
Dengue or dengue-like illnesses have been recorded in the Caribbean region as far back as the 17th century. 1–3 Yet, it is
only within the past few decades that this mosquito-borne fla-viviral disease has gained public health significance through-out the archipelago. 4,5 More than a million dengue cases have
been recorded for the region since the beginning of the 1960s, involving several thousand episodes of severe dengue hemor-rhagic fever (DHF) and hundreds of deaths. 2,6,7 It is noticeable
that dengue epidemics have recurred at narrowing intervals since the beginning of the 1980s, whereas outbreaks involving severe DHF have become regular phenomena for several of the larger island-states, such as Puerto Rico, the Dominican Republic, and Trinidad. 1,3,7–9 The manifestation of epidemic
DHF on these islands has coincided with the establishment of serotype co-circulation, thus resembling observations made in several South American countries during the same period. 9,10
It is generally accepted that hyperendemicity may increase the risk of epidemic DHF in a given population because of mecha-nisms of immunopathogenesis during heterotypic re-infection and/or escalation of strain virulence as a result of serotype competition. 11 The mechanisms of dengue pathogenesis
attrib-utable to viral factors or immunologic host responses are not fully described. 12–15 In addition, the actual population
thresh-old for continued hyperendemic transmission remains uncer-tain, as this is governed by a complex set of host, viral, and vector-dependent factors. 16–18
Studies of dengue activity in the Caribbean have for the most part been carried out on the larger islands of Puerto Rico, Cuba, and Trinidad where effective surveillance sys-tems and dedicated research groups have been established for several years. These studies have provided insight on the epidemiology and clinical manifestations, as well as genetic determinants of dengue infections in the sizeable insular populations (range 1.1–11.2 million). 19–26 However, of the 24
additional island-states or territories in the Caribbean, only the Dominican Republic, Haiti, and Jamaica hold popula-tions greater than one million. 27 The population base for the
remaining Caribbean islands ranges from less than 10,000 to
just over 400,000 people (average 143,000), while occupying no more than 10% of the total landmass of the archipelago. 27,28
Available dengue data from the smaller islands are very limited but do indicate a general pattern of single serotype outbreaks followed by periods of low or undetected transmission in pop-ulations of less than 200,000. 6,10,29 Given inadequate
surveil-lance and restricted sampling for most of the smaller islands it is difficult, however, to discern the true extent of serotype transmission. Likewise, there have been few studies assessing the epidemiology or clinical profile of dengue transmission within the smaller Caribbean populations. Hence, it remains uncertain whether the small-island setting of the Caribbean fall below the required population threshold for maintaining hyperendemic activity and to what extent this might affect the manifestation of disease severity at the population level.
The aim of this study was to explore the level of viral activ-ity, seroprevalence, and general epidemiology of dengue trans-mission in the small-island setting of the Caribbean using Grenada as an example. The prospective 18-month study involved the introduction of active, laboratory-based surveil-lance in an attempt to attain the highest possible detection rate for the clinical cases at all primary health care facilities on the island.
MATERIALS AND METHODS
Study area. Grenada is the main island of the tri-island
state of Grenada, Carriacou, and Petit Martinique. It is the most southerly (12.1°N, 61.4°W) of the Windward Islands within the Caribbean archipelago, with the nearest mainland country of Venezuela located 160 km to the south. The 310 km 2 island is of volcanic origin with relatively narrow coastal
plains and a mountainous interior. The tropical climate is tempered by the northeast trade winds, which carry a yearly average precipitation of more than 350 cm on the windward side of the island, and less than 150 cm on the lower coastal plains. Rainfall and humidity levels are highest between June and November, whereas average day temperatures remain constant throughout the year at 29°C (26–32°C). 30 Grenada
parishes each hold minor townships and villages, the majority of which are located along the coastline of the island.
Surveillance program. An active surveillance and diagnostic program was established from January 2001 through June 2002. The surveillance approach was adapted from the Guide-lines for Prevention and Control of Dengue and Dengue
Hemorrhagic Fever , published by the Pan American Health
Organization (PAHO). 31 All public and private primary health
care units (HCUs), i.e., medical offices, health clinics, medical stations, and hospitals, were invited to participate in the program via the Division of Epidemiology, Ministry of Health, and the Environment (MOH). Reminders of the program were issued regularly to all HCUs by the Public Health and Surveillance Nurse, who held the overall responsibility of actively identifying and studying all probable dengue cases on the island. The HCUs were requested to submit 5 mL venous blood and a standardized investigation form for all patients fitting the World Health Organization (WHO) criteria for probable dengue virus infection (i.e., acute febrile illness with two or more symptoms including, headache, retro-orbital pain, myalgia, arthralgia, rash, and hemorrhagic manifestations). 32
All patients with suspected dengue were included in the study regardless of age, sex, place of residence, and nationality.
Laboratory support. The diagnostic analyses were
con-ducted at the laboratory of the Windward Islands Research and Education Foundation (WINDREF) located at St. George’s University in Grenada. The Laboratory of Micro-biology at the General Hospital in St. George’s acted as the formal collection point for all blood samples referred to the WINDREF laboratory. To the extent possible, samples were also collected directly from the individual HCUs or delivered to the WINDREF laboratory by HCU staff. All HCUs were requested to notify the WINDREF laboratory immediately on blood draw to facilitate sample receipt and tracking. The requirement for rapid sample transfer and cold-chain delivery was emphasized throughout. Test results were reported directly to the referring HCU and to the Division of Epidemiology (MOH).
Detection of dengue virus. Acute phase serum (blood draw
< 6 days after symptom onset) was tested for the presence
of dengue virus using a semi-nested reverse transcription-polymerase chain reaction (RT-PCR) assay modified from Lanciotti and others. 33 A second convalescent phase sample
(> 14 days post onset) was requested for all RT-PCR negative
cases to confirm the result by paired serology (see later). The RT-PCR negative cases were reported as “indeterminate” if the second sample was unattainable.
Detection of anti-dengue IgM antibodies. Convalescent
serum (blood draw ≥ 6 days after symptom onset) were tested for the presence of anti-dengue IgM antibodies using either of two commercially available assays: INDX Dip-S-Ticks kit (PanBio INDX Inc., Baltimore, MD) or Dengue Duo IgM and IgG Rapid Strip Test (PanBio, Windsor, Australia ). Both assays were completed according to the instructions of the manufacturer. Samples positive for anti-dengue IgM antibodies were recorded as evidence of presumptive acute dengue virus infection (i.e., infection within the past 3 months). A second late-convalescent phase sample (blood draw > 14 days after
symptom onset) was requested for paired serology (see later), if an early convalescent phase sample (blood draw 6–10 days post onset) tested IgM negative. These cases were reported as “indeterminate” if a second sample was unattainable.
Detection of anti-dengue IgG antibodies. An IgG capture enzyme-linked immunosorbent assay (ELISA), modified from Chungue and others 34 and Kuno and others 35 was used for paired
acute and convalescent phase samples to confirm dengue virus infection by seroconversion (minimum 4-fold increase of anti-dengue IgG-ELISA titers between paired samples). Single acute phase samples with positive virology were also tested to differentiate between primary and secondary infections, as described by Miagostovich and others. 36
Phylogenetic analysis. A 600 nucleotide region in the Enve-lope gene was sequenced using the Taq DyeDeoxy Terminator Cycle Sequencing kit and an ABI PRISM 377 DNA sequencer (Applied Biosystems, Foster City, CA ). Sequence assembly and alignment were completed with Lasergene software (DNAStar, Madison, WI), whereas phylogenetic trees were generated using PHYLIP. 37 All sequencing and phylogenetic procedures
were kindly conducted by the Centers for Disease Control and Prevention (CDC) Dengue Branch in Puerto Rico.
Ethical approval. Formal ethical clearance and research
approval of the study were granted by the Grenada MOH, as well as the Internal Review Boards at WINDREF and St. George’s University, Grenada.
Data analyses. Data management and statistical analyses
(odds ratio [OR], with 95% confidence intervals [CIs]) were completed using Excel 2000 (Microsoft Office XP, Microsoft Corp., Redman, WA).
RESULTS
Temporal distribution of recorded cases. The active
dengue surveillance program recorded 545 cases of suspected dengue from January 2001 through June 2002 ( Figure 1 ). The distribution of cases was highly skewed with only 38 referred patients during the initial 9 months of the study against 507 cases during the latter 9 months. A short transition period characterized by a shift in reported age and increased frequency of suspected cases (monthly average of 14) occurred between October (Week 42) and December 2001 (Week 51). This pre-epidemic phase was followed by the onset of full pre-epidemic momentum with continuous weekly reporting of suspected cases lasting ~25 weeks and peaking with 139 cases during the month of March 2002.
Laboratory confirmation. The WINDREF laboratories
received a total of 577 blood samples during the surveillance period of which 427 (74%) were obtained from the patients in the acute phase (average of 2.3 days after symptom onset). Another 116 samples (20%) were obtained from the patients 6 days or more after symptom onset (average of 11.6 days) while a total of 34 samples had no record of status because of incomplete investigation forms. Of the 545 patients with suspected dengue, 223 were laboratory confirmed as either acute ( N = 150) or presumptive acute ( N = 73) dengue virus infections. Thirty suspected cases tested negative while 289 cases were categorized as “indeterminate,” and three were lost to testing. The causative serotype was identified by RT-PCR for 139 of the acute infections (3 DENV-2 and 136 DENV-3), whereas the infecting serotypes were unknown in 11 cases confirmed by seroconversion.
Age-specific disease rates. More than 71% (5/7) of the
cases during the pre-epidemic and epidemic period (October 2001–June 2002). The median age for confirmed cases shifted from 9 years (range 3–27 years) to 27 years (range 1 month–78 years) between the periods of low and high transmission.
Symptoms, age, and immune status. Classic dengue fever
(DF) was reported for 94% (198/211) of confirmed cases with recorded symptoms. Less than 6% (12/211) of the cases had DF with hemorrhagic manifestations such as ecchymosis, epitaxis, or hematuria, whereas a single patient was diagnosed with DHF, Grade II. Disease manifestations in children of verbal age (4–14 years) appeared somewhat milder than for adults, as these children were less likely to report symptoms such as headache, myalgia, and arthralgia ( Table 1 ). Hemorrhagic manifestations were not recorded for any child (< 15 years of
age) with confirmed or suspected dengue.
The immune status was assessed for 140 confirmed cases, of which 113 (81%) were identified as secondary infections. Stratification by age, for Grenadian nationals, showed previous infection in 83% (10/12) of children (< 15 years of age) as
com-pared with 95% (60/63) of adults. Of the 12 DF patients with hemorrhagic manifestations, five suffered secondary infections, whereas the status for the remaining seven is unknown. The single case of DHF occurred in a 30-year-old male with sec-ondary infection. A sub-analysis comparing the symptoms of primary and secondary cases did not reveal any significant dif-ference at the given sample size, except for a higher occurrence of rash (OR = 3.2, 95% CI = 1.4–7.6) and lower occurrence of joint pain in primary infections (OR = 0.4, 95% CI = 0.1–0.9).
Spatial and temporal disease distribution. The first labora-tory-confirmed DENV-3 infection, defined as the index case, occurred in November 2001. The parish of St. George was noted as the epicenter of the DENV-3 outbreak (December 2001– June 2002) hosting the index case and nearly 70% (117/171) of all confirmed cases with stated residence. The index case was a 48-year-old male residing ~5 km southeast of the town of St. George’s. He had no history of traveling before disease onset, but his unique profession as a boat surveyor and service provider to the yachting community would have involved daily inter-action with sailors having arrived recently from other dengue endemic Caribbean islands or the South American continent.
The highest number of confirmed cases during the DENV-3 epidemic was observed within the capital town of St. George’s and the immediate suburbs and bordering peri-urban com-munities. The confirmed disease rate for the entire parish of St. George (32:10,000) was twice that of neighboring St. David (16:10,000) presenting the second highest transmission level. St. Patrick located furthest to the North of the Capital par-ish, recorded the lowest rate of detected cases for the entire island (3:10,000) ( Figure 2 ). A lag period of 2 or more weeks was observed between the epidemic onset in St. George’s (Week 52, 2001) and the occurrence of confirmed cases in the remaining parishes. The main activity in these areas took place between Week 5 and Week 19 (2002), with less than four weekly cases reported from each parish.
Identification of circulating strains. Phylogenetic inference by comparison to reference and regional strains grouped a DENV-2 isolate from Grenada within the America/Asian
Figure 1. Temporal distribution of dengue cases recorded in Grenada from January 2001 through June 2002. Weekly numbers of suspected and laboratory-confirmed cases of dengue, according to the reported time of symptom onset. Three modes of disease dynamics were noted as low endemic, pre-epidemic, and acute epidemic transmission.
Table 1
Reported signs and symptoms
Symptom
4–14 years ≥ 15 years Association Positive no. (%) Positive no. (%) OR1* 95% CI†
Fever 35 (100) 156 (96) NA NA–NA
Headache 27 (77) 149 (92) 0.3 0.1–0.8
Retro-orbital pain 19 (54) 115 (71) 0.5 0.2–1.0
Myalgia 22 (63) 143 (88) 0.2 0.1–0.5
Arthralgia 13 (37) 132 (81) 0.1 0.1–0.3
Rash 7 (20) 26 (16) 1.3 0.5–3.5
Nausea/vomiting 13 (37) 80 (49) 0.6 0.3–1.3
Diarrhea 2 (6) 34 (21) 0.2 0.1–1.0
Chills 16 (46) 129 (80) 0.2 0.1–0.5
Cough 7 (20) 44 (27) 0.7 0.3–1.7
Sore throat 6 (17) 42 (26) 0.6 0.2–1.6
Nasal congestion 7 (20) 31 (19) 1.1 0.4–2.8
Petechia 0 – 4 (2) 0.0 NA–NA
Purpura/ecchymosis 0 – 3 (2) 0.0 NA–NA
Haematemesis 0 – 4 (2) 0.0 NA–NA
Melaena 0 – 4 (2) 0.0 NA–NA
Epistaxis 0 – 4 (2) 0.0 NA–NA
Bleeding gums 0 – 3 (2) 0.0 NA–NA
Haematuria 0 – 3 (2) 0.0 NA–NA
Vaginal bleeding 0 – 0 – NA NA–NA
Jaundice 0 – 2 (1) 0.0 NA–NA
Total cases 35 162
General manifestations reported for children of verbal age (4–14 years) and adult patients with confirmed dengue infection.
genotype, whereas isolates of DENV-3 were shown to belong to Genotype Group III.
Reported travel activity and temporary immigration.
Seven patients with confirmed infections reported regional or international traveling within 14 days from symptom onset. Serotype DENV-3 was reported before or during this particular period from each of the regional countries stated as the site of travel (Brazil, Dominica, Guyana, and Martinique). None of the specific cases could be linked, however, to the origin of the Grenadian outbreak, as each case occurred after the epidemic onset.
The number of monthly visitors arriving in Grenada from countries with confirmed DENV-3 activity was compared with the number of dengue cases detected on the island during the entire surveillance period. Highest travel activity was recorded in August 2001 (month of carnival festivities) with 4,026 stay-over arrivals against a monthly average of ~2,106 (Annual Statistical Reports 2001 and 2002. St. George’s, Grenada: Research and Planning Department, Grenada Board of Tourism). The peak in travel activity from known DENV-3 endemic countries pre-ceded the pre-epidemic phase with ~5 weeks and the detection of autochthonous DENV-3 transmission by nearly 10 weeks.
DISCUSSION
This study offers an insight to dengue transmission in the small-island setting of the Caribbean, by a combination of
virologic, immunologic, epidemiologic, and clinical data from the patients at primary health care facilities on the island of Grenada from January 2000.
Limitations of surveillance system. Underreporting of
clinical cases cannot be excluded, yet the study recorded as many as 546 cases of suspected dengue in just 18 months as compared with 220 cases during the previous 7 years (1994– 1999) of mandatory but passive surveillance. 6 The introduction
of active surveillance and rapid diagnostic feedback was widely credited for enhancing the index of suspicion and the report compliance among health care professionals on the island.
Limitations of diagnostic assay. The large number of “inde-terminate” acute phase cases ( N = 289) suggest inadequate performance by the RT-PCR assay in use. This was confirmed by comparison of the assay to paired serology (IgG ELISA) based on 30 available paired samples. The comparison revealed an overall RT-PCR specificity of 100% (12/12), but a sensitivity of merely 50% (9/18). Optimization of the RT-PCR assay was not attempted. However, it is noted that the DENV-3 specific primer was set at a lower concentration than specified in the original protocol. This may have prevented sufficient amplification in specimen with limited viral numbers as assumed for secondary infections with high IgG titers. Furthermore, a comparison of the Grenadian DENV-3 sequence and that of the DENV-3 specific primer (TS3: based on prototype H87) revealed two nucleotide substitutions at the 5′-end of the primer, flanking a “weak” motif of double Adenine bases. 33 The inadequate sequence homology may
have caused incomplete primer annealing, exacerbating the negative impact of low primer concentration.
Disease transmission patterns and viral activity. Over the course of 18 months, we observed three distinct dengue trans-mission patterns in Grenada, in the form of low endemic, pre-epidemic, and acute epidemic activity. The period of low endemicity involved a monthly average of just four suspected cases mainly within the pediatric population and with confirmed infections ( N = 7) occurring up to 12 weeks apart. Unfortunately, the causative serotype(s) was not identified for any of these cases. However, DENV-2 (American Asian genotype) was strongly suspected given confirmed activity in the years leading up to this study (1997–2000) and in addition to three cases identified during the epidemic period. 10,29 The
pre-epidemic phase was characterized by an increase in sus-pected cases to a monthly average of 14, with confirmed cases ( N = 12) occurring no more than 2 weeks apart. The phase also carried a noticeable shift toward the adult demographic, while including the first laboratory-confirmed DENV-3 infection in Grenada (index case). The overall changes in transmission dynamics and the shift in the age span of confirmed cases, as observed during the pre-epidemic phase, suggested a loss of age-dependant herd immunity with the replacement of DENV-2 by DENV-3—effectively predicting an oncoming epidemic. The DENV-3 epidemic was the largest dengue outbreak yet to be recorded on the island.
Age, disease manifestations, and immune status. The
highest levels of DENV-3 activity were recorded for ages 15 to 44 years, which could reflect increased exposure among the more active age groups of the population. Yet, presentation bias resulting from milder manifestations in children cannot be excluded (see later). It is also probable that DENV-3 was transmitted on the island as part of a regional epidemic in 1963–1965. 3,38 Hence, DENV-3 immunity acquired during the
1960s could be a contributing factor to the lower disease rates observed for Grenadians over 40 years of age.
The vast majority of patients in this study reported symp-toms of DF, whereas more than 90% of the Grenadian patients were shown to suffer a secondary infection. Unfortunately, the Tourniquet test and platelet counts were not performed systematically for all patients. This may explain the lower rates of observed hemorrhagic phenomena and leakage, as compared with other reports from the region. However, it is remarkable that hemorrhagic phenomena were reported exclusively for adults during the epidemic period, and with no significant association to the immune status of these patients. Importantly, there were no reports of hemorrhagic manifes-tations among children experiencing secondary infections—in fact, pre-exposed children, of verbal age, reported generally milder manifestations than did adults. Although based on relatively few cases, the lower incidence of severe manifesta-tion in the pediatric populamanifesta-tion seems to fit a general pattern in South America and the Caribbean. 39–41 At the global scale,
however, children with secondary infections continue to suffer the brunt of morbidity and mortality due to DHF and dengue shock syndrome. 42–44
Spatial and temporal disease distribution. Patients with
laboratory-confirmed dengue and stated area of residence ( N = 178) were clearly concentrated in the urban south—an area broadly defined by the political boundaries of St. George and St. David. These parishes hold 50% (48,543/96,551) of the island population, but accounted for 79% (140/178) of all confirmed cases with reported residence during the surveillance period. A systematic assessment of spatial or temporal vector indices and climate parameters was not included in this study. Yet, provisional entomologic data indicated high infestation levels of Aedes aegypti throughout the study period for small villages, townships, and urban areas alike, with average House and Breteau indices at 14% and 23%, respectively (Vector Control Division, MOH, unpublished data). This would imply that most Grenadians were at high risk of vector exposure, irrespective of location or time of year. In fact, it is noted that the epidemic occurred during the dry season.
Possible disease surveillance biases caused by geographic differences in patient presentation and report compliance by the different HCUs cannot be excluded for this study. Yet, the spatial distribution and apparent delay in viral progression into the sparser populated, rural communities fit the general understanding that dengue virus transmission is favored by urban conditions, such as high densities of susceptible hosts and possibly higher activity levels of urban Ae. aegypti to that of its rural counterpart. 45,46 A shift toward increased rural
trans-mission has, however, been noted in several dengue endemic countries and could also take place in Grenada. 47 As tourism
replaces the agricultural-based economy, employment oppor-tunities are shifting from the rural areas to the tourism centers in and around St. George’s. The resulting boost in rural–urban commuting could increase the possibilities for viral spread into the rural communities.
Possible transmission clusters. Infection within private households was suggested in at least 20 cases. Yet, the true extent of household transmission remains uncertain, as does the level of transmission at other typical sites of host aggregation. Notably, four medical doctors working at the General Hospital (GH) in St. George’s had confirmed dengue infection. Three of the doctors reported symptom onset during the first
month of acute epidemic transmission (January 2001). The observation suggests that the 240-bed, unscreened hospital building may have acted as a focal point of transmission. Indeed, the presence of Ae. aegypti was confirmed on the hospital compound at the time of the study (Vector Control Division, MOH, unpublished data). Unfortunately, it was not possible to determine the overall incidence of dengue infections among hospital personnel, or to determine whether infections occurred in patients who were hospitalized because of unrelated disease. Furthermore, available data do not reveal the actual distribution between hospitalized as opposed to ambulant dengue patients referred from the GH. It is notable however that the Accident and Emergency ward was credited for the majority of suspected (98/139) cases reported from this particular HCU.
St. George’s University located about 4 km to the south of the capital town was recognized as another cluster of activ-ity, as the university clinic referred more than 8% (18/223) of all confirmed infections. These data align with a previous study showing an annual infection rate of 5% within the pre-dominantly foreign student body of ~2,000 people (Pulim S, unpublished data). Non-nationals, many of which were tem-porary visitors from non-endemic areas, represented an additional 25 confirmed infections (11.3%). The relatively high rate of student and other non-national cases could well reflect differences in the health care seeking behavior to that of Grenadians. However, with an influx of 261,392 stay-over visitors during the study period, non-nationals should be recognized as a potential source of susceptible hosts capa-ble of extending endemic transmission and intensifying epi-demic activity (Annual Statistical Reports 2001 and 2002. St. George’s, Grenada: Research and Planning Department, Grenada Board of Tourism).
Possible viral entry and spread of dengue viruses. All
DENV-3 strains isolated from Grenadian patients were shown to belong to Group III, the only genotype of DENV-3 known to circulate in the region at the time of the study. The genotype was introduced into Central America from South Asia in 1994, possibly via Africa. 48 Its presence in the Caribbean Basin was
confirmed in 1998, with concurrent reports from Jamaica, St. Kitts and Nevis, Barbados and Aruba. 29 By the end of year 2000,
DENV-3 had been isolated from all of the larger and almost half of the smaller Caribbean islands. 29 The 2001 appearance
of DENV-3 in Grenada followed a continuation of regional epidemics in 2000 and 2001 including, for the latter year, 13,000 reported cases in Cuba; 4,471 in Martinique; 1,043 in Barbados; 292 in St. Lucia; and 2,244 in Trinidad and Tobago. 6,41 Nearby
Venezuela reported 83,180 dengue cases while Brazil reached a toll of more than 400,000. 6 Apart from the smaller island of
St. Lucia (population = 153,000), all outbreaks were associated with at least two serotypes, even as the main epidemic activity was attributed to DENV-3. 6 Noticeably, DHF cases were
reported from all outbreaks except for St. Lucia. 6
DENV-3 endemic countries, doubled from the monthly aver-age. The combined increase in potential DENV-3 carriers and the assembly of large volumes of carnival revelers (locals and visitors) would seem to provide favorable conditions for viral introduction and subsequent proliferation.
General epidemiology. This study identified the fourth
serotype to be confirmed in Grenada, within a 20-year span. Data recorded before this study are extremely limited but do suggest that the introduction of each serotype has been followed by epidemic transmission possibly quelled by increasing herd-immunity and/or replacement by a novel serotype. 10,29 Whereas the 2002 DENV-3 epidemic affected all
age groups (age range of confirmed cases: 1 month to 78 years), future dengue outbreaks may be limited to the age cohorts born after the last introduction or epidemic manifestation of each serotype (i.e., DENV-4: 1981, DENV-1: 1995, DENV-2: 1999, DENV-3: 2002). 3,38,49 Restrained by fewer numbers of
susceptible hosts, prospective epidemics may be of shorter duration but not necessarily of less severe character, given an increased risk of immunopathogenesis caused by heterotypic re-infection. Apparently, the Grenadian population may be too small to sustain hyperendemic transmission and thereby local escalation of strain virulence leading to severe outbreaks. However, there is a continuous risk that virulent strains may be imported from the larger, neighboring populations that suffer hyperendemic activity with frequent outbreaks (e.g., in Venezuela and Trinidad). Available data indicate that most of the small-island states in the Caribbean may have similar transmission and disease patterns as those observed in Grenada. Together, these islands offer an intriguing setting for the study of outbreak severity and the possible interactions between population size, generation of virulence, and immunopathogenesis.
There is no immediate prospect of dengue eradication, given the lack of available vaccines or sustainable methods for vec-tor elimination. As such, current efforts are focused on limiting dengue morbidity and mortality by curbing and possibly pre-venting epidemic transmission through vector control. 31 This
study indicates that active laboratory-based surveillance may identify pre-epidemic transmission in the small-island setting by early detection of novel serotypes, intensified disease activ-ity and/or a shift in affected age groups. The predictive capabil-ity of the active surveillance system should enable activation of timely and targeted prevention efforts. However, it is obvi-ous that such efforts must be tailored to the noted transmis-sion dynamics and risk factors of viral introduction and spread within the insular population. For Grenada, this includes a deliberate focus on the high-risk areas and population groups in the urban South with specific vigilance around periods of increased visitor influx from other dengue endemic areas.
Received July 7, 2008. Accepted for publication April 20, 2009.
Acknowledgments: We acknowledge the support of the Grenada Ministry of Health and the Environment and in particular Agnes Banfield, Public Health and Surveillance Nurse at the Division of Epidemiology without whom this project would not have succeeded. We are also grateful for the technical support, viral specimen, and laboratory reagents provided by Vance Vorndam, former Director of the Dengue Laboratory, CDC Dengue Branch in Puerto Rico. Finally, we thank Flemming Konradsen and Michael Alifrangis from the Department of International Health, Immunology, and Microbiology, University of Copenhagen for their comments and advice in prepara-tion of this manuscript.
Financial Support: Financial support was provided by the Danish Research Academy and the Windward Islands Research and Education Foundation. The American Society of Tropical Medicine and Hygiene (ASTMH) assisted with publication expenses.
Authors’ addresses: Karin L. Schiøler, Department of International Health, Immunology and Microbiology, University of Copenhagen, CSS, Øster Farimagsgade 5, DK-1014 Copenhagen K, Denmark, Tel: (+45)35327181, Fax: (+45)35327736, E-mail: ksch@sund.ku.dk. Calum N. Macpherson, Windward Islands Research and Education Foundation, True Blue Campus, St. George’s University, St. George’s, Grenada, Tel: (+473)4443997, Fax: (+473)4443041, E-mail: cmacpher son@sgu.edu.
Reprint requests: Karin L. Schiøler, Department of International Health, Immunology and Microbiology, University of Copenhagen, CSS, Øster Farimagsgade 5, DK-1014 Copenhagen K, Denmark, E-mail: ksch@sund.ku.dk.
REFERENCES
1. Gubler DJ, 1997. Dengue and dengue hemorrhagic fever: its his-tory and resurgence as a global health problem. Gubler DJ, Kuno G, eds. Dengue and Dengue Hemorrhagic Fever . First edi-tion. Wallingford, UK: CAB International, 1–22.
2. Ehrenkranz NJ, Ventura AK, Cuadrado RR, Pond WL, Porter JE, 1971. Pandemic dengue in Caribbean countries and the south-ern United States–past, present and potential problems. N Engl J Med 285: 1460–1469.
3. Gratz NG, Knudsen AB, 1996. The Rise and Spread of Dengue, Dengue Haemorrhagic Fever and Its Vectors: A Historical Review (Up to 1995) . Geneva: World Health Organization. 4. Hospedales CJ, 1990. Dengue fever in the Caribbean. West Indian
Med J 39: 59–62.
5. Caribbean Commission on Health and Development, 2006. National Library of Jamaica Cataloguing in Publication Data: Pan American Health Organization (PAHO) and Caribbean Community Secretariat (CARICOM).
6. World Health Organization, DengueNet. The DengueNet data-base and geographic information system, Available at: http:// www.who.int/csr/disease/dengue/denguenet/en/index.html. Accessed March 21, 2009.
7. Pinheiro FP, 1989. Dengue in the Americas, 1980–1987. Epidemiol Bull 10: 1–8.
8. Gubler DJ, Trent DW, 1993. Emergence of epidemic dengue/ dengue hemorrhagic fever as a public health problem in the Americas. Infect Agents Dis 2: 383–393.
9. Guzman MG, Kouri G, 2003. Dengue and dengue hemorrhagic fever in the Americas: lessons and challenges. J Clin Virol 27: 1–13. 10. Campione-Piccardo J, Ruben M, Vaughan H, Morris-Glasgow V,
2003. Dengue viruses in the Caribbean. Twenty years of dengue virus isolates from the Caribbean Epidemiology Centre. West Indian Med J 52: 191–198.
11. Gubler DJ, 1998. The global pandemic of dengue/dengue hemor-rhagic fever: current status and prospects for the future. Ann Acad Med Singapore 27: 227–234.
12. Kurane I, 2007. Dengue hemorrhagic fever with special emphasis on immunopathogenesis. Comp Immunol Microbiol Infect Dis 30: 329–340.
13. Kurane I, Ennis FA, 1997. Immunopathogenesis of dengue virus infections. Gubler DJ, Kuno G, eds. Dengue and Dengue Hemorrhagic Fever . First edition. Wallingford, UK: Cab International, 273–290.
14. Pang T, Cardosa MJ, Guzman MG, 2007. Of cascades and perfect storms: the immunopathogenesis of dengue haemorrhagic fever-dengue shock syndrome (DHF/DSS). Immunol Cell Biol 85: 43–45.
15. Green S, Rothman A, 2006. Immunopathological mechanisms in dengue and dengue hemorrhagic fever. Curr Opin Infect Dis 19: 429–436.
16. Kuno G, 1997. Factors influencing the transmission of dengue viruses. Gubler DJ, Kuno G, eds. Dengue and Dengue Hemorrhagic Fever . First edition. Wallingford, UK: Cab International, 61–88. 17. Kuno G, 1995. Review of the factors modulating dengue
18. Anderson JR, Rico-Hesse R, 2006. Aedes aegypti vectorial capac-ity is determined by the infecting genotype of dengue virus. Am J Trop Med Hyg 75: 886–892.
19. Alvarez M, Rodriguez-Roche R, Bernardo L, Vazquez S, Morier L, Gonzalez D, Castro O, Kouri G, Halstead SB, Guzman MG, 2006. Dengue hemorrhagic fever caused by sequential dengue 1–3 virus infections over a long time interval: Havana epidemic, 2001–2002. Am J Trop Med Hyg 75: 1113–1117.
20. Guzman MG, Kouri GP, Bravo J, Soler M, Vazquez S, Morier L, 1990. Dengue hemorrhagic fever in Cuba, 1981: a retrospective seroepidemiologic study. Am J Trop Med Hyg 42: 179–184. 21. Bravo JR, Guzman MG, Kouri GP, 1987. Why dengue
haemor-rhagic fever in Cuba? 1. Individual risk factors for dengue hae-morrhagic fever/dengue shock syndrome (DHF/DSS). Trans R Soc Trop Med Hyg 81: 816–820.
22. Rigau-Perez JG, Ayala-Lopez A, Garcia-Rivera EJ, Hudson SM, Vorndam V, Reiter P, Cano MP, Clark GG, 2002. The reappear-ance of dengue-3 and a subsequent dengue-4 and dengue-1 epi-demic in Puerto Rico in 1998. Am J Trop Med Hyg 67: 355–362. 23. Rigau-Pérez JG, Ayuso-Lamadrid A, Wolff DR, Reiter P, Kuno G, The Puerto Rico Association of Epidemiologists, 1994. Dengue severity throughout seasonal changes in incidence in Puerto Rico, 1989–1992. Am J Trop Med Hyg 51: 408–415.
24. Morens DM, Rigau-Perez JG, Lopez-Correa RH, Moore CG, Ruiz-Tiben EE, Sather GE, Chiriboga J, Eliason DA, Casta-Velez A, Woodall JP, 1986. Dengue in Puerto Rico, 1977: public health response to characterize and control an epidemic of multiple serotypes. Am J Trop Med Hyg 35: 197–211.
25. Bennett SN, Holmes EC, Chirivella M, Rodriguez DM, Beltran M, Vorndam V, Gubler DJ, McMillan WO, 2003. Selection-driven evolution of emergent dengue virus. Mol Biol Evol 20: 1650–1658.
26. Rodriguez-Roche R, Alvarez M, Gritsun T, Halstead S, Kouri G, Gould EA, Guzman MG, 2005. Virus evolution during a severe dengue epidemic in Cuba, 1997. Virology 334: 154–159. 27. Central Intelligence Agency, The 2008 World Factbook, U.S.
Government. Available at: http://www.cia.gov/cia/publications/ factbook/index.html. Accessed March 21, 2009.
28. Encyclopc´dia Britannica, 2008. Greater Antilles. Encyclopedia Britannica Online. Available at: http://www.britannica.com/eb/ article-9037895. Accessed March 21, 2009.
29. Pan American Health Organization, Number of Reported Cases of Dengue and Dengue Hemorrhagic Fever (DHF), Region of the Americas. Dengue Surveillance: Incidence/Mortality/Case Fatality, Available at: www.paho.org/english/ad/dpc/cd/dengue. htm. Accessed March 21, 2009.
30. Meditz SW, Hanratty DM, 1987. Caribbean Islands: A Country Study (Grenada) . Washington, DC: GPO for the Library of Congress, Federal Research Division of the Library of Con-gress. Available at: http://countrystudies.us/caribbean-islands/. Ac cessed March 21, 2009.
31. Pan American Health Organization, 1994. Dengue and Dengue Hemorrhagic Fever in the Americas: Guidelines for Prevention and Control . Washington, DC: Pan American Health Orga-nization (PAHO).
32. World Health Organization, 1997. Dengue Haemorrhagic Fever: Diagnosis, Treatment, Prevention and Control . Geneva, Switzer-land: World Health Organization (WHO).
33. Lanciotti RS, Calisher CH, Gubler DJ, Chang GJ, Vorndam AV, 1992. Rapid detection and typing of dengue viruses from
clini-cal samples by using reverse transcriptase-polymerase chain reaction. J Clin Microbiol 30: 545–551.
34. Chungue E, Marche G, Plichart R, Boutin JP, Roux J, 1989. Comparison of immunoglobulin G enzyme-linked immunosor-bent assay (IgG-ELISA) and haemaglutination inhibition (HI) test for the detection of dengue antibodies. Prevalence of den-gue IgG-ELISA antibodies in Tahiti. Trans R Soc Trop Med Hyg 83: 708–711.
35. Kuno G, Gomez I, Gubler DJ, 1991. An ELISA procedure for the diagnosis of dengue infections. J Virol Methods 33: 101–113. 36. Miagostovich MP, Nogueira RMR, dos Santos FB, Schatzmayr
HG, Araujo ESM, Vorndam V, 1999. Evaluation of an IgG enzyme-linked immunosorbent assay for dengue diagnosis. J Clin Virol 14: 183–189.
37. Felsenstein J, 1993.PHYLIP [Phylogeny Inference Package], ver-sion 3.5c. Seattle, WA: University of Washington.
38. Pan American Health Organization, The History of Dengue and Dengue Hemorrhagic Fever (DHF) in the Region of the Americas, 1635–2001: A Timeline for Dengue in the Americas . Pan Ameri can Health Organization: Division of Disease Prevention and Control. Available at: http://www.paho.org/ English/AD/DPC/CD/dengue_history.htm. Accessed March 21, 2009.
39. Rahman M, Rahman K, Siddque AK, Shoma S, Kamal AH, Ali KS, Nisaluk A, Breiman RF, 2002. First outbreak of dengue hemor-rhagic fever, Bangladesh. Emerg Infect Dis 8: 738–740. 40. Guilarde AO, Turchi MD, Jr JB, Feres VC, Rocha B, Levi JE, Souza
VA, Boas LS, Pannuti CS, Martelli CM, 2008. Dengue and den-gue hemorrhagic fever among adults: clinical outcomes related to viremia, serotypes, and antibody response. J Infect Dis 197: 817–824.
41. Pelaez O, Guzman MG, Kouri G, Perez R, San Martin JL, Vazquez S, Rosario D, Mora R, Quintana I, Bisset J, Cancio R, Masa AM, Castro O, Gonzalez D, Avila LC, Rodriguez R, Alvarez M, Pelegrino JL, Bernardo L, Prado I, 2004. Dengue 3 epidemic, Havana, 2001. Emerg Infect Dis 10: 719–722.
42. World Health Organization, 2009. Fact sheet No. 117: Dengue and dengue haemorrhagic fever. Available at: http://www.who.int/ mediacentre/factsheets/fs117/en/index.html. Accessed March 21, 2009.
43. Torres JR, Torres-Viera JM, García HSJR, Baddour Y, Bajares A, Castro JM, 2004. Prognostic factors of clinical outcome in non-paediatric patients with dengue haemorrhagic fever/dengue shock syndrome. Dengue Bull 28: 68–76.
44. Lee MS, Hwang KP, Chen TC, Lu PL, Chen TP, 2006. Clinical char-acteristics of dengue and dengue hemorrhagic fever in a medi-cal center of southern Taiwan during the 2002 epidemic. J Microbiol Immunol Infect 39: 121–129.
45. Gubler DJ, 2004. Cities spawn epidemic dengue viruses. Nat Med 10: 129–130.
46. Chadee DD, Martinez R, 2000. Landing periodicity of Aedes aegypti with implications for dengue transmission in Trinidad, West Indies. J Vector Ecol 25: 158–163.
47. Guha-Sapir D, Schimmer B, 2005. Dengue fever: new paradigms for a changing epidemiology. Emerg Themes Epidemiol 2: 1. 48. Lanciotti RS, Lewis JG, Gubler DJ, Trent DW, 1994. Molecular
evolution and epidemiology of dengue-3 viruses. J Gen Virol 75: 65–75.
