Epidemiological, virological, and entomological characteristics of dengue from 1978 to 2009 in Guangzhou, China

Lei Luo, Huiying Liang, Hu YuShan, Weijia Liu, Yulin Wang, Qinlong Jing, Zhicong Yang

Journal of Vector Ecology · 2012 · 57 citations · 32 references

DOIFull text

Open access

Concepts

Abstract

To understand its unprecedented resurgence, we examined the epidemiological, virological, and entomological features of dengue in Guangzhou during 1978–2009. Cases reported to the Guangzhou Centre for Disease Control and Prevention and data from virological and entomological surveillance were analyzed from three periods: 1978–1988, 1989–1999, and 2000–2009. Although cases decreased over time: 6,649 (1978–1988) to 6,479 (1989–1999) to 2,526 (2000–2009), geographical expansion resulted in districts with an average incidence >2.5/100,000, increasing from five (1978–1988, 1989–1999) to seven (2000–2009). Age distribution (mean age: 34.9 years) provided a trend of increasing dengue incidence among adults, and there was a significantly higher incidence among men with a sex ratio of 1.15:1 (P<0.001). Cases occurred from May through November with a peak between August and October, and a long-term trend was characterized by a three to five-year cyclical pattern. The most frequently isolated serotypes were DENV-2 (1978–1988) and DENV-1 (1989–1999 and 2000–2009). Seasonal fluctuations in immature densities of Aedes albopictus (sole transmission vector in Guangzhou) were consistent with the dengue seasonality. After a 30-year apparent absence, DENV-3 had reemerged in 2009. The current epidemiological situation is highly conducive to periodic dengue resurgences. Thus, a high degree of surveillance and strict control measures in source reduction should be maintained. Dengue fever (DF), caused by a group of four closely related but antigenically distinct flaviviruses (DENV-1 through DENV-4), is the most important arthropod-borne viral disease in tropical and subtropical countries/regions. The clinical spectrum of dengue infection ranges from asymptomatic, undifferentiated fever, dengue hemorrhagic fever (DHF), to dengue shock syndrome (DSS), which is frequently fatal because of abnormal capillary permeability and plasma leakage. Worldwide it accounts for approximately 2.5 billion people at risk, 100 million dengue infections and 24,000 dengue-related deaths each year (Gubler 2002b). Recently, it has become a major international public health concern. Shepard et al. (2004) estimated a loss of 420 disability-adjusted life years (DALYs) per 1,000,000 per year, comparing to that of meningitis (390 DALYs) and twice that of hepatitis in the region of southeast Asia. In Latin America and the Caribbean, authors estimated an average loss of 658 DALYs per 1,000,000 per year, with the same order of magnitude as tuberculosis (Meltzer et al. 1998). The Disease Control Priorities Project estimated the global burden of dengue as 528,000 DALYs for the year 2001, corresponding to a burden of 264 DALYs per 1,000,000 per year for 2.5 billion people living in or traveling to tropical and subtropical areas at risk of dengue (Cattand et al. 2006). There had been no evidence of dengue outbreaks in China's mainland during 1949–1977. However, a dramatic geographic expansion of epidemic DF from southeast Asian countries west to India, Sri Lanka, the Maldives, and Pakistan and east to China (Yang et al. 2009) have been seen in the past two decades. An outbreak of dengue in Foshan city (contiguous to Guangzhou) of Guangdong province in 1978 started the endemic of dengue in southern China, including Guangdong, Guangxi, Hainan, Taiwan, Fujian, and Zhejiang (Wen et al. 1998). In August 1978, autochthonous dengue infections first spread to Guangzhou from Foshan. Since then, highly localized and relatively sporadic yearly outbreaks have been observed in Guangzhou. The cumulative total of dengue cases in Guangzhou accounted for nearly 66.9% of those in Guangdong and 55.6% in China. Although there is a perceived increase in disease burden, DF epidemiology in Guangzhou has not been well documented. Here, we report the year-to-year variability of DF in Guangzhou from 1978 to 2009 to characterize its epidemiological, virological, and entomological dynamics and provide proper information essential for planning and development of relevant control/preventive measures against dengue. Located in southern China on the Pearl River, and adjacent to Hong Kong, Macao, Taiwan, and southeast Asia, Guangzhou is the transportation, industrial, financial, and trade center of South China. As of 2009, the census population was about 11.64 million, with a population density of 1,627 persons/km2 in the metropolitan area. Spanning from 112°57′ to 114°03′ E longitude and 22°26′ to 23°56′ N latitude, Guangzhou is characterized by a humid subtropical climate influenced by the Asian monsoon. The annual average temperature in Guangzhou is 22.6° C (72.7° F), and the annual rainfall is over 1,700 mm with a relative humidity of 66% to 70%. Both natural and social conditions in Guangzhou are conducive to mosquito growth, reproduction, and the transmission of dengue. Dengue fever is a legally notifiable infectious disease in China. Monthly notified dengue fever cases in the metropolitan area of Guangzhou City from 1978 to 2009 were retrieved from the Notifiable Infectious Disease Report System (NIDRS) at the Guangzhou Center for Disease Control and Prevention (Guangzhou CDC). The information required for each notification includes demographic data such as name, unique identification number, age, ethnic group, gender, residential and school or workplace addresses, dates of symptom onset and diagnosis, and whether the diagnosis was clinical or confirmed by laboratory tests. Only physician-diagnosed (clinical diagnosis) or laboratory-confirmed cases of dengue, according to the World Health Organization's criteria, were included in this study. All imported cases of the disease were removed. The study was approved by the Institutional Review Board of the Guangzhou CDC. After obtaining informed consent, acute- and convalescent-phase blood samples of DF suspected cases were drawn and used to detect positive IgG and IgM with dengue virus-specific immunochromatographic test (ICT) and enzyme-linked immunosorbent assay (ELISA) (Parida et al. 2001, Wu et al. 1997). Sera collected during day 1–7 after the onset of symptoms are referred to as acute-phase sera. Early and late convalescent sera refer to the specimens collected during day 8–13 and day 14–30, respectively. Dengue virus was further isolated by inoculating the sera from the confirmed cases of dengue patients reported to the Viral Immunology Section in the Laboratory of Viral Diseases in the Guangzhou CDC, into C6/36 mosquito cell cultures (Yamada et al. 2002). The virus was then identified and its E (envelope) gene sequence was amplified by Reverse transcription polymerase chain reaction (RT-PCR) to construct the phylogenetic tree of the local DF agents (Carvalho et al. 2010). To amplify the entire sequence of the DENV E gene, we designed four serotype-specific primer pairs for RT–PCR with Primer Express version 3.0 (Applied Biosystems, Foster City, CA, U.S.A.), as shown in Table 1. A confirmed case of dengue virus infection was defined as febrile illness associated with: (i) the isolation of viral RNA; (ii) positive test of real-time one-step RT-PCR; (iii) positive seroconversion or ≥ four-fold increase in dengue-specific IgM or IgG antibody from appropriately timed paired serum; or (iv) high-titer dengue-specific IgM and IgG antibody in a single serum specimen where cross-reaction to Japanese encephalitis (JE) had been excluded. A sentinel system with 24 representative sites randomly scattered through the rural, peri-urban, and urban areas has been used in Guangzhou since early 1978. However, the indicators used in surveilance were very limited at first. Except for the information about the detection and monitoring of adult vectors, surveillance for larval populations was unavailable from 1978 to 2001. Adult mosquitoes have been sampled with an aspirator and flash torch since 1978. The pipette method in tire dumps and a small stainless steel dipper with iron handle were used for the sampling of Aedes larvae from permanent or temporary water containers around the 24 sentinel sites. The samples of larvae from different habitats were brought to the laboratory for a reconfirmation of species identity. The sampling of larvae and adults was carried out weekly to observe the seasonal changes in both larval and adult densities. The average monthly container indices and adult index were then estimated. Conventional Aedes larval indices (Breteau index-BI, House Index-HI, and Container Index-CI systematically started from 2002) and (Mosquito or Oviposition Index-MOI systematically started from 2005) and Man-Hour-Density (MHD) were used to describe the risk of DF transmission (Ansari and Razdan 1998). The indices are calculated as follows: BI=number of positive containers per 100 houses; HI=percentage of positive houses divided by the total inspected houses; CI=percentage of positive containers divided by the total inspected containers; MOI=the number of captured mosquitoes or mosquito eggs divided by the number of mosquito oviposition devices; MHD (bites/man/hour)=n×60/txp (where n=no. of mosquitoes; t=time in min; p=no. of persons). To investigate the dynamic spatial clustering and dispersion of dengue incidence in Guangzhou, China, three time periods were considered according to the epidemiologic characteristics: 1978 through 1988 (introduction, mainly from southeast Asia countries to urban areas), 1989 through 1999 (introduction, from overseas to pre-urban areas), and 2000 through 2009 (indefinite sources of infection throughout the city). Population data from the Guangzhou Municipal Bureau of Statistics was used to calculate incidence and case fatality rates for each study time period. Analysis of dengue reports over time was based on monthly/yearly dengue incidence. Analyses of annual cycles and seasonal pattern were based on monthly incidence aggregated over all years of data included in the study. We used χ2 or the Fisher exact test as appropriate with two-side significance level of 0.05. All analyses were performed using Epi Info™ Version 3.5.1 (CDC, Atlanta, GA, U.S.A.). The total dengue cases reported in Guangzhou was 6,649 (8.75/100,000) during 1978–1988, 6,479 (8.68/100,000) during 1989–1999, and 2,526 (5.75/100,000) during 2000–2009. Similarly, the number of DHF/DSS cases decreased over time from 170 (0.22/100,000) during 1978–1988, 54 (0.07/100,000) during 1989–1999, and 11 (0.03/100,000) during 2000–2009. The DHF/DSS cases as a percentage of total dengue cases also decreased from 2.56% to 0.83% to 0.44%. Of these, 12 deaths were directly attributed to dengue infection, which resulted in a case fatality rate of 0.08% for all dengue cases and 5.1% for DHF/DSS cases. The diagnosis was typically made clinically, on the basis of reported symptoms and physical examination; this applied especially in endemic years. Acute- and convalescent-phase serum samples were collected from the 2,127 DF suspected cases during 1978–1988, 2,769 during 1989–1999, and 1,271 during 2000–2009. The corresponding figure of samples serologically positive for dengue infection according to established diagnostic criteria were 88.15%, 63.84%, and 77.66%, respectively. The mean age of Guangzhou indigenous dengue cases was 34.9 years. Overall age distribution of cases showed that incidence increased with age, peaked among the 21– to 30-year-old age group and then gradually declined (Table 2). However, the highest incidence of DF during 2000–2009 was among 31– to 40-year-olds (11.06/100,000). The incidence rate among ≤ 10 year olds decreased dramatically over time from 10.68/100,000 to 5.84/100,000 to 2.68/100,000. In general, DHF/DSS incidences by age group were conincident with those observed for DF, but all deaths occurred in people older than 50 years. Case reports also indicated that men were more frequently infected with dengue than were women (7.19 per 100,000 vs 6.98 per 100,000; χ2=42.59, P<0.001) with a male to female ratio ranging from 1.18 during 1978–1988 to 1.16 during 1989–1999 and 1.09 during 2000–2009. However, of the 235 DHF/DSS cases in this study, 150 (63.83%) were female. From 1978 to 2009, there were five epidemic years in which the number of annual dengue cases reached more than 1,000, i.e., 3,395 cases in 1980, 1,197 cases in 1985, 1,445 cases in 1986, 5,683 cases in 1995, and 1,476 cases in 2002, with four additional small outbreaks in 1987 with 586 cases, in 1990 with 459 cases, in 1991 with 305 cases, and in 2006 with 859 cases. During the study period, the NIDRS reported between 0 and 5,683 dengue cases per year corresponding to a mean incidence of 7.04 per 100,000 populations, with the case fatality rate ranging from 0.00 to 0.98%. From 1978 through 1988, nine deaths were reported compared with three deaths during 1989–1999. No deaths were reported in Guangzhou during 2000–2009. cycles were observed three to five years with an increase of DHF/DSS cases during epidemic years. rate per 100,000 of reported dengue cases and the adult Aedes density in Guangzhou, 1978–2009. the distribution of dengue cases during the study period. There were and seasonal features in the of indigenous cases during epidemic and years. In a increase in reported cases of dengue was seen at of May and peaked in late and then gradually in October, with no cases reported from the of to the of of the outbreak years of 1980, 1985, 1986, 1995, 2002, and 2006 showed but cases were for and a high and peak occurred from August through The highest of dengue occurred between and fluctuations in the number of dengue cases in the Guangzhou surveillance system during 1978–2009. the distribution of the total number of indigenous cases in 12 districts in the surveillance system from 1978 to 2009. showed that the of dengue cases were in urban areas and in areas and sporadic cases occurred in and dengue incidence per 100,000 and of sentinel surveillance sites by districts in Guangzhou, 1978–2009. The number of districts with an average incidence than increased from five during 1978–1988 and 1989–1999, to seven during 2000–2009. the of the there had been no districts an of indigenous The highest were reported in the urban areas and during 1978–1988 and but during 1989–1999, by the areas the highest number of dengue cases, districts After for population in each the urban incidence was than areas and during 1978–1988 and 2000–2009 In the study period, acute-phase serum samples positive for such as and were to of the samples were for dengue with DENV-1 by DENV-2 and DENV-3 Although all four dengue virus serotypes had been there was epidemiological in the DENV serotypes (Table those that from the or epidemic years in 1980, 1985, 1986, 1995, 2002, and 2006). The reported most frequently during 1978–1988 was pattern during 1989–1999 and 2000–2009 DENV-1 was the most frequently reported However, after years of apparent DENV-3 had reemerged in 2009, which had for the years. during epidemic year, a single with the of during which DENV-1 and after the of DENV-1 showed that four were identified in Guangzhou, and DENV-1 have been from southeast Asia countries such as and DENV-3 there were two with the from the and the with the and two with the from and the with the to and albopictus were the species in Guangzhou, and no positive case was The density of albopictus was about the of a control for four in and the adult Aedes density decreased from during to than for the However, the outbreak in the of Guangzhou occurred in 1995, and further of Aedes control as by the adult Aedes Aedes Aedes and Aedes after 2001, two occurred in and The adult Aedes density was around per from to There not to be between the adult Aedes density and the incidence of dengue cases. also showed no between annual larval density index and and the dengue incidence after 2001. The monthly average of started from started from and MHD are in Both the adult density and the of Aedes were observed in throughout the the peak in with a density of per by the density in per and the density in per increasing by from to year with the increasing of and Similarly, the highest and were also in and were very during the Seasonal larval indices and densities of Aedes in Guangzhou. collected over three by the Guangzhou surveillance system to the epidemiological features from a of patients with dengue. The incidence of dengue in Guangzhou is estimated to be from 0 to per 100,000 As shown in study, dengue cases in Guangzhou older age year age which is to the that dengue is a et al. et al. et al. et al. 2010). Although evidence of increasing dengue incidence among older age in both Latin America and southeast Asia were et al. et al. et al. 2001, and there is also a trend infections in age in countries such as and et al. et al. et al. are not but be related to the the level of and social development and and et al. the of different virus or Aedes et al. and a in and In the adult and in age pattern adults be caused by the that most of the cases are infections have to to be infected However, the in epidemiologic pattern of dengue cases as an to health to to provide for high risk with DF to the of cases. A of in infection rates and of disease is important for dengue control and In with the distribution in the region of the in which higher of cases and higher incidence rates in compared with were dengue in Guangzhou most with a male to female ratio of However, more is that are in the 235 DHF/DSS cases in this As early as study out that among disease but for more illness be related to an antibody et al. it is that in China, the incidence among women be a that is related to at to infection, and of Thus, and both and social are to of the outbreak data that a major outbreak occurred in Guangzhou in a cyclical pattern of approximately three to five years. the had become during However, the is that the most outbreak occurred in the dengue case in during the between and and the of the peak in are The dengue incidence and have been associated with the 2001, et al. However, the of different entomological in this pattern have not been further to be climate are to be dengue cases had been in the urban areas and where density and of the vector conditions for increased However, showed that the geographical of notified dengue cases has significantly in Guangzhou over the three to urban areas and The geographical and expansion of dengue incidence be attributed to the of risk in the in Guangzhou, including the of and social such as increasing and et al. Population data that in population density of the areas during the study be of the The density has been since the in 1978 and has to 2009. is by from in China during 1978–1988, and also from the metropolitan area during 1989–1999 and 2000–2009. changes also the disease However, infection rates to those in urban areas further in that an epidemic occurred in in in more than cases that were to be caused mainly by imported cases from countries et al. 2002, et al. et al. 2002). Although all four dengue virus serotypes were in this region throughout the study period, it is that a single was isolated from local dengue infection cases in all but be related to the that the outbreaks in Guangzhou were caused mainly by imported cases from countries and the and mosquito control measures the spread of imported In dengue infection in Guangzhou was caused by However, the in the from which had been for at years to DENV-3 in 2009, a to health because of those to the that the DENV-3 virus has been associated with a of such as DHF/DSS et al. and the in the has been related to outbreaks in countries et al. Thus, a number of additional including vector control and public with should be as as the among dengue virus and disease become more and more Aedes with and especially entomological was the vector in the areas of the tropical latitude, albopictus has a area of distribution in China, from to the southern of the et al. in the transmission of dengue, it has not been to be a major vector and is in a vector than (Gubler 1998). the of the are to as a of of dengue infections et al. 2002). During the study, albopictus was the species which was and most patients were as dengue infection cases, which at to the for case fatality and compared to both Latin America and southeast Asia et al. et al. Although that the density of adult Aedes is a in dengue incidence et al. we not observe a between the adult and larval Aedes density and incidence of dengue. be to the that the albopictus on blood in the day time (Yang et al. the surveillance of Aedes started mainly in the late However, in to the annual adult and larval density monthly fluctuations in immature and the seasonal pattern of the incidence of dengue during both epidemic and years. study on surveillance data to describe the epidemiology and the disease which have both and and of diagnostic were required to all dengue cases, cases with and not be and the high to ratio of also to et al. et al. epidemiological analyses of dengue in Guangzhou were mainly based on which are on the clinical of dengue infection be very which this diagnosis to However, an epidemic it is Thus, the disease burden be during but during (Gubler the and shown in age distribution are it is to the small and of information of where the samples all acute-phase serum samples positive for were to from serologically confirmed cases, the number of specimens for virus isolation was relatively also to a single was isolated in all but indicators and of entomological surveillance were not collected systematically during be the of the study are and to dengue in Guangzhou. In no data the and such a In with the of international population and global climate increasing dengue has a major public health to Guangzhou. epidemiological, virological, and entomological with an of the disease burden, as well as an vector control the to dengue transmission in Guangzhou. study was by the of China and the Guangzhou Municipal Bureau We are to of the South for in The authors that have no

References

32