Laboratory Surveillance of Dengue in Rio Grande do Sul, Brazil, from 2007 to 2013

Background According to official records, dengue was introduced in Brazil in the 80's; since then several epidemics have occurred. Meanwhile, in Rio Grande do Sul (RS, Southern Brazil) the first autochthonous case occurred only in 2007. Methodology and Principal Findings In this study we report laboratory surveillance of dengue cases and seasonality of positive cases, describe serotypes and characterize the epidemiological pattern of dengue in RS from 2007 to 2013. A total of 9,779 serum samples from patients with suspected dengue fever were collected and submitted to molecular and/or serological analyses for dengue virus identification and serotyping, based on viral isolation, NS1 antigen detection and qRT-PCR, or Dengue IgM capture ELISA and MAC-ELISA. The first autochthonous dengue case in RS was confirmed in 2007 (DENV-3). While in 2008 and 2009 only imported cases were registered, autochthonous infection waves have been occurring since 2010. The highest number of dengue infections occurred in 2010, with DENV-1 and DENV-2 outbreaks in Northwestern RS. In 2011, another DENV-1 and DENV-2 outbreak occurred in the Northwestern region; moreover, DENV-4 was detected in travelers. In 2012, DENV-1 and DENV-4 co-circulated. DENV-2 circulation was only detected again in 2013, in high frequency (56.7%), co-circulating with DENV-4 (35%). Most infections occur in adults during summer. Differences in prevalence between genders were observed in 2007 (60% females), 2008 (60.8% males) and 2009 (77.5% males). Conclusions According to results of dengue surveillance, there was an increase in the number of dengue cases in RS and of cities infested with Aedes aegypti, possibly as a consequence of introduction of new serotypes and the difficulty of health programs to control the vector.


Introduction
Dengue fever (DF) is the most important arthropod-borne viral disease and is considered a serious public health problem [1], affecting between 50-100 million people/year worldwide, causing 250,000 cases of Dengue Hemorrhagic Fever (DHF). Moreover, 2.5 billion people live in dengue-endemic countries, including Brazil [2].
The clinical response to infection by dengue virus (DENV) causes a spectrum of disease symptoms [3], [4], from mild DF to severe DHF and Dengue Shock Syndrome (DSS). DENV is transmitted by mosquitoes of the genus Aedes and consists of four closely related but antigenically distinct serotypes (DENV-1, DENV-2, DENV-3 and DENV-4) [4]. All four serotypes can be distinguished by serological and molecular methods [5], [6].
In Brazil, there are reports of dengue epidemics in 1916 and 1923, in the States of São Paulo and Rio de Janeiro (Southeast Brazil), respectively. However, the first laboratory-confirmed DENV case was reported in the State of Roraima (Northern Brazil) only in 1981-1982, with the introduction of DENV-1 and DENV-4, probably an extension of the DENV epidemics waves that had occurred in Central America and in northern parts of South America [1].
In 1986-1987, DENV-1 caused an outbreak in Rio de Janeiro and then spread to other Brazilian states; in 1990, DENV-2 was also introduced in Rio de Janeiro [1], [7]. During the 90's the vector Ae. aegypti spread throughout the national territory and led to dissemination of DENV-1 and DENV-2 to 20 of the 27 Brazilian States [7].
The first dengue cases caused by DENV-3 were reported in 2000 in Rio de Janeiro, and in 2001 in Roraima. Subsequently, it was identified in the entire country, except in the Southern states of Santa Catarina and Rio Grande do Sul (RS) [1].
Since its first introduction in the country in the late 80's, DENV-4 was identified again in Brazil only in 2008, in the city of Manaus, State of Amazonas (Northern Brazil) [8]. In the following years, DENV-4 spread nationwide and started to circulate together with the other serotypes in different regions of the country.
Today, all four serotypes circulate in Brazil, and dengue is a mandatory notifiable disease. Historically, the State of RS has registered dengue circulation since 1996, with imported cases of people coming from other areas [9]. Subsequent autochthonous cases were detected from January 2007 to May 2013, with dengue outbreaks caused by DENV-1, DENV-2 and DENV-4 [10][11][12][13].
We report here the results of Dengue laboratory surveillance since the first autochthonous case identified in January 2007 until May 2013 in RS. This is the first time-series report on dengue epidemiology in RS, Brazil.

Ethics Statement
This study has been approved by the Ethics Committee of IPB-LACEN/RS (Ethics Statement n. 371.278) and of UFCSPA

Patients and Samples
This study included human serum samples sent to the Central Laboratory of Public Health (IPB-LACEN/RS) for dengue surveillance diagnosis, including all patients with clinically suspected DF in RS. Clinical DF was based on the following symptoms: sudden onset of fever, headache, retro-orbital pain, body aches, nausea and vomiting, joint pains, weakness, and rash [4]. For each patient, the Dengue Notification Form -containing information such as demographic characteristics, date of notification, timing of sampling, symptoms, travel history, results of laboratory tests -was filled out at time of sample collection. Samples were analyzed using commercially available tests (NS1 antigen detection and Dengue IgM capture ELISA), and in-house IgM antibody capture ELISA (MAC-ELISA) as well as real time reverse transcription-polymerase chain (qRT-PCR) and viral isolation, depending on the days of onset of symptoms, as described below.
Dengue-positive serum samples that had been previously analyzed and serotyped in our laboratory were used as positive controls in qRT-PCR and MAC-ELISA assays.

Dengue Assays
The Brazilian National Dengue Program states that samples collected within five days of the onset of symptoms must be analyzed either by the NS1 antigen detection assay followed by qRT-PCR for confirmation (when NS1 antigen is positive), or directly submitted to viral isolation for confirmation and DENV serotyping. On the other hand, samples that are collected after six or more days of the onset of symptoms must be analyzed based on the serological method using the Dengue IgM Capture enzymelinked immunosorbent assay (ELISA); positive samples are then confirmed by MAC-ELISA [14]. Accordingly, only results obtained either with viral isolation, qRT-PCR and/or MAC-ELISA are valid for confirming dengue diagnosis.
The laboratory surveillance of all suspected dengue cases in RS is performed in IPB-LACEN/RS. From 2007 to 2010, IgM-based assays were performed at IPB-LACEN/RS and at the Adolfo Lutz Institute (IAL); serotyping based on viral isolation [13] was done at IAL. Since 2011, IPB-LACEN/RS has been in charge of all analyses, including DENV detection and serotyping, as well as serological tests, according to the methodologies described below.
NS1 Antigen Detection. NS1 antigen detection was performed with the Platelia TM Dengue NS1 Ag kit (BIO-RAD Laboratories, France), according to the manufacturer's instructions.
DENV Serotyping by qRT-PCR. Dengue virus serotyping was based on qRT-PCR as described by Johnson et al. (2005) [15]. Total viral RNA was extracted from 200 mL of NS1-reagent serum specimens using Pure Link Viral RNA/DNA Mini Kit (Invitrogen-Life Technologies, USA). In singleplex reaction mixtures, 7 mL of viral RNA was combined with 25 pmol of each Antibody Tests (IgM). For serological tests, the Dengue IgM capture ELISA kit (Panbio-Inverness Medical Innovations, Australia) was used, according to manufacturer's instructions. Positive samples were confirmed using MAC-ELISA according to Kuno et al. [16]. Antigens for this test were provided by Evandro Chagas Institute, State of Pará.

Statistical Analyses
Descriptive statistics for the study population characteristics and laboratory findings were performed using SPSS Inc. (Chicago, IL, Version 17.0). X 2 test was applied for comparisons among proportions (standardized adjusted residuals $2 was used when the number of the cases observed was greater than the expected) and p,0.05 was considered significant.  (Table 1).

Results
In the state of Rio Grande do Sul dengue infection cases occurred during summer, displaying a peak in March and April, while few cases are reported from July to November (winter and spring) and these are often imported cases; the number of cases starts to increase again in January and February. The monthly distribution of dengue-positive cases along the years (2007-2013), in RS, is shown in Figure 1.
In 2007, a total of 1,529 dengue suspected cases were notified and 15.6% (238) were positive based on MAC-ELISA and viral isolation ( Table 1). The frequency was similar in the age groups analyzed (Table 2), and higher in female than in male (143 and 95, respectively, Figure 2). The DENV-3 serotype was detected in that year.
During 2008 only imported cases were reported and computed. Of the 917 notifications, 10.6% were dengue positive, with higher prevalence in the age group under 10 years old; the frequency of infection was higher among males (60.8%). There were only imported cases in that year; the few samples had been collected with more than five days after the onset of symptoms, thus serotyping was not possible.
In 2009, only imported dengue cases were computed as well, with fewer notifications (297) than in 2008. Positivity was 13.5%, the most affected gender was male (77.5%), and no difference among age groups was found. There is no information about the serotypes circulating in 2009 because collected samples were inappropriate to viral isolation.
In 2010, a total of 2,345 suspected cases of dengue were reported in RS, among which 41.7% were autochthonous (375/

Discussion
According to official records, dengue virus was introduced in Brazil in the 80's, and since then, all serotypes have been circulating in different regions of the country. In RS, Southernmost State of Brazil, the first imported dengue cases were notified in 1996, and the first autochthonous cases were confirmed in 2007 in cities located in Northwestern RS (Giruá, Horizontina, Tucurundi, Três de Maio) and Northern RS (Erechim) [17]. Since, autochthonous dengue transmission has occurred in RS, as it has been well documented in other regions where Aedes aegypti mosquitoes are present.
In the years that RS registered autochthonous cases, difference between male and female was observed only in 2007, when the prevalence was more frequent in females (60%); on the other hand, in 2008 and 2009, when all cases were imported, dengue was more frequent in males (60.8% and 77.5%, respectively), probably associated to more mobility of adult men. A previous study based on phylogenetic analysis of DENV-4 that circulated in RS and in the State of São Paulo in 2011, showed that the Brazilian strains are closely related to strains that have been circulating since 1981, when DENV-4 was first introduced in South America, suggesting that the virus has gone through recent evolution for at least 4 to 6 years [13]. DENV-4 may have penetrated the Brazilian population earlier than 2010, causing milder disease; therefore, and also due to a higher prevalence of DENV-1 and DENV-2, the virus could have been present but not detected. The reemergence of DENV-4 evidences that the replacement of a dominant circulating genotype is associated with the rising of a previously rare lineage [13].
The highest number of notifications and many autochthonous cases occurred in 2010, the year with the highest number of positive dengue infections (38.4%), compared to other years when the frequency did not exceed 17.5%. In 2011, 276 infections occurred and 18% were in Porto Alegre. Even though there were fewer notifications in 2012, dengue infections increased by 23.5% in Porto Alegre. Thereafter, the number of cities with autochthonous dengue cases increased, from five cities in 2007 to 53 in 2013.
Even though the number of dengue cases in RS increased in the last years, it was proportionally lower than in other regions of Brazil [19]. This condition may be associated with the climate of the region, with rigorous winters that are not suitable for mosquitoes reproduction, different from the rest of Brazil where temperatures are high year-around [3]. Control of DENV spread relies primarily on reduction of vector populations, which is usually done by elimination of still water sources and use of insecticides. Despite the effort of the State Program for Surveillance of Ae. aegypti, detection of the vector increased in RS [20], confirming the persistence of infested cities and the emergence of new occurrences. In 2007 there were 52 cities infested with Ae. aegypti, while in May 2013 this number has risen to 116 (496 cities), making of RS an endemic region.
Molecular studies are also fundamental to understand viral evolution, viral dynamics and its interaction with hosts, and to predict the severity of the disease. DENV phylogenetics and phylodynamics in RS should be taken into account in future studies. This approach will contribute to improve the quality of surveillance data, to predict circulation of different dengue serotypes and genotypes as well as to evaluate the impact of preventive measures such as mosquito control or a future vaccine.
Finally, considering that dengue was recently introduced in RS, the control of further outbreaks requires a network composed of well-trained and skilled health professionals, entomologists, environment surveillance agents, laboratory technicians and molecular biologists, as well as social education and governmental support.