Residual Spray for the Control of Aedes Vectors in Dengue Outbreak Residential Areas

The incidence of dengue in Malaysia has shown an increasing trend since the year 2000. Vector control is the primary approach for dengue control in Malaysia. There is an urgent need for new or modified approaches such as the residual spraying on the outer walls that can potentially last long enough to control the Aedes population, particularly in the outbreak-prone areas. In this field study, we conducted outdoor residual spraying (ORS) using a newly formulated polymer-enhanced suspension concentrate (SC-PE) of deltamethrin. The objectives of this study were to evaluate the efficacy of ORS using deltamethrin SC-PE and its effect on wild Aedes populations and to assess its residual bio-efficacy on painted cement walls against the pyrethroid-susceptible strains of laboratory-reared Aedes mosquitoes. Three rounds of spraying in a four-month cycle were conducted between 2014 and 2015 in four residential areas (low-rise and high-rise housing types) in Hulu Langat, Selangor. The bio-efficacy of the insecticide was evaluated by assessing its impact on vector population using ovitrap surveillance. Standard WHO wall deposit bioassay was adapted to determine bio-efficacy of deltamethrin, i.e. post 30 min knockdown and post 24 h mortality after exposure. During the treatment period, we observed significant reductions in the population of Ae. albopictus in the sprayed low-rise housing in both semi-indoor and outdoor environments, while in the high-rise housing, there was also a significant decline in Ae. aeHow to cite this paper: Hamid, N.A., Noor, S.N.M., Saadatian-Elahi, M., Isa, N.R., Rodzay, R.M., Ruslan, B.M., Omar, T., Norsham, M.I.M., Amanzuri, N.H., Khalil, N.A., Zambari, I.F., Kassim, M.A.M., Zaman, M.K.K., Effendi, A.M.B., Hafisool, A.A., Peng, L.T., Poong, B., Ibrahim, M., Roslan, N.A. and Lim, L.H. (2019) Residual Spray for the Control of Aedes Vectors in Dengue Outbreak Residential Areas. Advances in Entomology, 7, 105-123. https://doi.org/10.4236/ae.2019.74009 Received: August 5, 2019 Accepted: October 14, 2019 Published: October 17, 2019 Copyright © 2019 by author(s) and Scientific Research Publishing Inc. This work is licensed under the Creative Commons Attribution International License (CC BY 4.0). http://creativecommons.org/licenses/by/4.0/ Open Access


Introduction
Dengue is the predominant mosquito-borne disease in humans, caused by one of the four serotypes of dengue virus (DENV-1 -4) that belongs to the genus Flavivirus. The virus is transmitted through the bite of an infected female of sympatric Aedes aegypti and Ae. albopictus. Dengue disease is endemic in more than 120 countries with the highest disease burden reported in Asia and Latin America [1] [2]. Within Asia, Malaysia ranked the third most affected country between the years 1991-2007 [3]. The incidence of dengue in Malaysia showed an escalating trend from 31.6 per 100,000 population in 2000 to 159.7 per 100,000 population in 2010 [4]. Dengue infection affected all ethnicities and gender with an observed predominance in males and Malays [4] [5].
In the absence of an effective and safe vaccine against the dengue virus, vector control remains as the primary approach for the prevention and control of dengue. Interruption of dengue transmission through vector control in the Caribbean, America, Cuba, and Singapore has resulted in a substantial reduction of the mosquito population and human cases of dengue [6] [7] [8] [9] [10]. Unfortunately, these success histories did not sustain and have failed to prevent dengue outbreaks due to several reasons, including inappropriate implementation, lack of sustainability over-time and reduced herd immunity [10] [11] [12] [13].
These are the compelling reasons for the continuation of research and development on new insecticides and tools for dengue control. At present, there is no consensus regarding the type of vector control approach that can bring the largest impact on dengue transmission for both epidemic situation and sustained management [14]. Hence, there is an urgent need for new or modified approaches such as the residual spraying on the outer walls that can potentially last long enough to control the Aedes population, especially in the outbreak areas. Indoor residual spraying (IRS) involves the application of long-lasting chemical insecticides on the inner walls, eaves and ceiling of properties where the mosquitoes might encounter the insecticides while resting on these structures [15]. IRS was historically used as the primary tool in the Global Malaria Eradication Program (1955)(1956)(1957)(1958)(1959)(1960)(1961)(1962)(1963)(1964)(1965)(1966)(1967)(1968)(1969), which has resulted in the total elimination of malaria in 37 out of 143 countries while reducing malaria burden in other countries [15] [16] [17] [18]. A few studies have shown that the IRS is a promising dengue control tool and has high efficiency in reducing the Aedes populations. The application of IRS using deltamethrin in Peruvian Amazon has appeared to produce >80% mortalities in Ae. aegypti in bioassay test for up to eight weeks post-spraying when applied on wood, unpainted wood and brick, and reduced adult index from 18.5 to 3.1, a month after the intervention [19]. Research in Merida, Mexico showed that treating inner house surfaces with bendiocarb has reduced Ae. aegypti infestation by 60% in three months period [20], and targeted IRS research in Cairns, Australia demonstrated that IRS along with location-based contact tracing had decreased the likelihood of dengue cases by 86% -96% in sprayed premises in comparison to unsprayed controls [21].
Deltamethrin polymer-enhanced suspension concentrate (SC-PE) (K-Othrine  Polyzone) is classified as pyrethroid and is one of the WHO recommended insecticides for IRS against malaria vector [22]. The insecticide is a new formulation of existing deltamethrin that has a prolonged residual effect through its property of the proprietary polymer. The polymer layer releases deltamethrin particle over an extended period and protects the deltamethrin particle from rainfall and abrasion. Consequently, less spraying applications are needed that in turn improve the cost-effectiveness of the residual spraying method and mini- The objectives of this study were to evaluate the efficacy of ORS using deltamethrin SC-PE and its effect on wild Aedes populations and to assess its residual bio-efficacy on painted cement walls against the pyrethroid-susceptible strains of laboratory-reared Aedes mosquitoes.

Study Site and Design
This study was conducted in four residential neighbourhoods in Hulu Langat District, Selangor. The study sites were selected based on epidemiological and ovitrap surveys by the Health District Office that has classified these zones as dengue outbreak hotspots with high vector density of Aedes spp. (≥10% ovitrap index threshold based on MOH guidelines). Two accommodation types were included, i.e. low-rise housing (landed properties with one to three-story design) and high-rise housing (non-landed properties having four stories and above).
We chose low-rise housing of terrace houses (Seksyen 2, Bandar Rinching, (MOH) guidelines shall be followed whereby activities under the mosquito control program that includes source reduction, larviciding, and space spraying will be conducted within 200 m radius of the infected house.

Insecticide Formulation and Application
The insecticide used was polymer-enhanced deltamethrin (K-Othrine  Polyzone, Bayer Crop Science, Monheim am Rhein, Germany), which contains 62.5 g/L deltamethrin as its active ingredient and formulated as a suspension concentrate (SC). The insecticide was prepared as recommended by the manufacturer and was sprayed using a calibrated hand compression sprayer (Hudson X-Pert, Chicago, IL, USA) with a flat fan nozzle (Teejet 8002, Spray Systems Co., Bessemer, AL, USA). Some of the external walls were not sheltered walls and were exposed to rainfalls. We decided to use a larger concentration (30 mg/m 2 ) than the WHO-recommended dosage for IRS in malaria control (20 -25 mg/m 2 ) [22] to offset the amount of insecticide applied, withstanding the dry and wet season in Malaysia. All accessible external walls in the house front yards and backyards were sprayed by trained staffs following procedures from WHO [28]. Safety procedures when spraying and handling of the insecticide were also followed [28]. The areas with aquarium or fish ponds, edible plants, surfaces that were near to the electric supply, water filter and water container were avoided. House owners were advised to keep their pets away during the spraying activities until the insecticide has dried completely. The percentage coverage of houses that have been sprayed in both types of accommodation was as described in Table 1. The coverage is defined by the total number of houses sprayed during a spray round divide by the total number of houses that were targeted [15]. The ORS coverage for terrace houses was not 100% due to inaccessibility to the houses since the homeowners were not at home at the time of spraying.

Ovitrap Surveillance
The number of larvae which translated to population densities of Ae. aegypti and Ae. albopictus was assessed in the control and treated areas via ovitrap surveillance. We performed ovitrapping as described in Lee [29]. Ovitraps (300 ml black plastic containers, diameter 7.0 cm and height 9.0 cm) filled with three-quarter of tap water along with the rough surface of the oviposition paddles (made from hardboard, approximately 10 cm × 2.5 cm × 0.3 cm) facing upwards were placed at the ground level in both semi-indoor and outdoor environments. A total of 60 ovitraps and 40 ovitraps were placed in randomly selected houses in the low-rise and high-rise housing, respectively. The number of ovitraps was equally distributed between the semi-indoor and outdoor environments. Semi-indoor is defined as the area outside of the house units but shielded by the roof, e.g. the porch and hallway. The ovitraps were collected after seven days and immediately transported to the Institute for Medical Research (IMR) laboratory. New ovitraps were placed at the same time when exposed ovitraps were collected. Any disturbances to the exposed ovitraps, e.g. stolen or vandalized, ovitraps were tipped over, invasion by insects were recorded. The recovered ovitrap contents along with the paddle were poured into labelled plastic containers and separated according to the study sites and date of collection. Beef liver powder (Difco Laboratories, MD, USA) was added to the ovitrap contents. Fourth instar larvae were identified and separated according to species by using established taxonomy keys [30] [31] under a compound microscope (Nikon Eclipse  E100, Japan). Contents were examined for species identification until no newly emerged larvae were found.

WHO Wall Deposit Bioassay
The residual activity of deltamethrin SC-PE was evaluated following standard WHO wall deposit bioassay procedures with a few modifications [32]. The susceptible laboratory-reared strains of Ae. aegypti and Ae. albopictus from insectary of Medical Entomology Unit, IMR were used in this bioassay. Wall bioassay was conducted biweekly until week 16 in each cycle and was performed on the treated wall surfaces of approximately five percent of the total houses. The houses for the bioassays were randomly selected. Standard transparent plastic cones (internal diameter 8.5 cm and internal height 5.5 cm) were fixed vertically onto the wall surfaces. Three bioassay cones, each containing 10 laboratory-reared mosquitoes (sucrose-fed, 2 -5 days old) were prepared for each Aedes species.
Mosquitoes were introduced gently into the bioassay cone using battery-operated aspirator. The aperture of the bioassay cone was plugged with a cotton ball. Mosquito knockdown was observed at the 1-minute interval for 30 min. After the exposure, the mosquitoes were aspirated out and transferred into clean paper cups. The mosquitoes were fed with 10% (w/v) sugar solution with the holding condition of 28˚C ± 2˚C and 75% ± 10% relative humidity. The mortality was recorded at 24 h post-exposure. The classification of adult mosquitoes as knocked down or dead was as described by WHO [33].

Meteorological Data
We set up a weather station in each study site. Temperature and relative humidity were recorded using thermohygrometer Tinytag View 2 (TV-4500, Gemini Data Loggers Ltd., Sussex, UK) while rain gauge (RGR126N, Wireless Rain Gauge, Oregon Scientific, Oregon, US) was used to closely monitor rainfalls.

Data Analysis
Ovitrap index (OI) was calculated to assess the density of the Aedes population. was used to compare the OI and mean number of larvae between unsprayed control and treated sites.
( ) where S 1 = sum of the ranks assigned to the sample observations from Population 1, S 2 = sum of the ranks assigned to the sample observations from Population 2, n = number of individuals in the sample [34]. Bioassay results of 30 min knockdown (KD30) and mortality rates were calculated and analysed according to [35]. The association between time (months) and mortality were analysed using log-time probit regression [35] [36]. Comparisons of mean mortality between the types of housing were analysed using Mann-Whitney Test as described above.
All analyses were performed using the Statistical Package for Social Sciences

Surveillance of the Aedes Population
The mean temperature recorded during the entire period studied was 29.29˚C ± 0.117˚C (26.38˚C -32.20˚C) while the mean relative humidity was 73.45% ± 1.004% (65.46% -81.44%). The mean precipitation recorded by the rain gauge was the highest during cycle 2 with 6.46 ± 1.089 mm precipitations in the low-rise residences and 7.47 ± 1.406 mm in the high-rise residences ( Table 2).
The rain amount was not statistically significant between the different housing types during the treatment period ( Table 2).
Comparisons of the OI between unsprayed control and treated low-rise residences for semi-indoor ovitraps showed that there were no statistical differences (p-value > 0.05) during the treatment period (Table 3).  Ae. albopictus where the differences were statistically significant (p = 0.000) between control and treated area (Table 4). Altogether, these results suggested that ORS with deltamethrin SC-PE was reducing Ae. albopictus larvae number, indicating a decreased of Ae. albopictus density in the low-rise residences.
The OI of semi-indoor ovitraps at treated high-rise residences declined after the first round of spraying with a significant difference between unsprayed control and treated site ( Table 5). The OI subsequently increased in cycle 2 and decreased in cycle 3 ( Table 5). As opposed to the low-rise housing, we found that the ORS is effective in reducing Ae. aegypti abundance as evident from the significantly lower number of Ae. aegypti larvae in the treated site than the unsprayed control (   ovitrap after spraying (Table 5). Both indices of OI and number of larvae for outdoor ovitraps in the high-rise residences did not differ considerably between unsprayed control and treated site ( Table 6).

Residual Bio-Efficacy of Deltamethrin SC-PE on Cement Walls
Wall bioassays done on cement-painted walls in terrace houses displayed considerable variation in KD30 between the three spray cycles (Figure 1)   rates in cycle 1 and 3 remained above 80% for 12 weeks in exposed Ae. aegypti, while for 6 weeks in cycle 2 ( Figure 1). The KD30 values, however, did not appear to be reflected in the percentage mortality where cycle 2 exhibited high mortality rate (above the WHO 80% mortality threshold) until week 14 for both Aedes spp., as opposed to cycle 1 and 3 that lasted until week 12 ( Figure 2).
Overall, there was a downward trend of KD30 and mortality from week 2 to week 16, suggesting that the residual bio-efficacy of deltamethrin SC-PE was reducing over time.
Like low-rise housing type, variation in knockdown rates was also reported in the high-rise residences. The overall trend showed that same descending trend   above 80% until week 12 in cycle 3 ( Figure 3). There were very high mortality rates (above 80% threshold) observed for both Aedes spp. in cycle 3, which lasted until week 16 ( Figure 4). Looking at the mortality rates between the different accommodation types suggests that deltamethrin SC-PE has a higher residual efficacy in eliminating Aedes laboratory strains in the apartment (>80% for 16 weeks) compared to terrace houses (>80% for 14 weeks).
The time (months) needed to eliminate 50% of the Aedes population (LT 50 ) and 90% of the population (LT 90 ) of exposed mosquito was analyzed using log-time probit regression. The results of probit analysis revealed that Ae. aegypti exposed to deltamethrin SC-PE in terrace houses exhibited LT 50 of 22.32 months and LT 90 of 6.61 months (Table 6). In comparison, longer periods of LT 50 of 34.44 months and LT 90 of 9.52 months were seen for Ae. aegypti in the treated apartment ( Table 7), suggesting that the residual bio-efficacy in apartment exceeded that of terrace houses. Ae. albopictus followed the same trend, at which treated apartment displayed a longer period of LT 90 (6.26 months) compared to terrace houses (LT 90 = 4.49 months) ( Table 7). Post-hoc analyses supported probit regression results demonstrating that treated apartment has higher means of mortality rate for both Aedes spp. in comparison to treated terrace houses (Table 8).

Discussion
This was the first field study that evaluated the efficacy of deltamethrin SC-PE in an ORS intervention.  [38]. The study, however, was only carried out in a short period of 12 weeks. Our study used the improved formulation of deltamethrin and showed an impact on reducing Aedes population size.
The initial dominance of Ae. albopictus during the baseline in the semi-indoor and outdoor environments in low-rise residences could be elucidated through its behavior. It is worthwhile mentioning that in terrace houses, semi-indoor ovitraps were placed outside of houses, in which they were partly shielded by roofs.
Ae. albopictus is largely considered as exophilic (outdoor resting) and exophagic (feeds outdoor), which breeds in natural habitats such as tree holes as well as man-made containers [ [42]. ORS using deltamethrin SC-PE appeared to produce a quantifiable impact on Ae. albopictus population size, but not on Ae. aegypti, depicted from the significant reduction in the number of larvae and the shifting of dominance to Ae. aegypti at post-spraying. We think that ORS may have affected Ae. aegypti abundance but the method that we used was not sensitive enough to capture this impact since Ae. aegypti was not the dominant species in this type of housing. Resistance to deltamethrin seemed unlikely since standard WHO susceptibility assay performed using another pyrethroid insecticide of permethrin 0.75% against wild Ae. aegypti in the treated area exhibited 100% mortality [43].
In the high-rise housing type, Ae. aegypti dominated in the semi-indoor environment during the baseline. The apartment semi-indoor was mostly sheltered walls that were nonetheless considered as indoor, and so the initial abundance of Ae. aegypti was highly anticipated due to its behavior. Ae. aegypti is primarily known as endophilic species (indoor resting), endophagy (feeds indoors) and prefers indoor artificial containers for breeding [26] [27]. We discovered that the number of Ae. aegypti population size decreased in the treated site after spraying and that the dominance has changed to Ae. albopictus in cycle 1 and 3, suggesting ORS was effective in eliminating Ae. aegypti. ORS did not seem to have an effect on the outdoor environment. A likely explanation was that the outdoor ovitraps were placed quite far from the sprayed walls, offering a less residual effect of the ORS.
Several external factors could have influenced the OI and the number of larvae per ovitrap in both control and treated sites. A likely factor was the dengue control and preventive actions performed by the Health District Office if there is a reported dengue case in the areas, which might influence the Aedes population.
Another factor that should be taken into consideration was the rainy season in Malaysia. During the second cycle, we observed an increase in OI in both types abundance. The evaluation of efficacy was based solely on semi-indoor and outdoor ovitrap collection, which did not consider the probability of indoor cryptic breeding sites such as the toilet tanks [44].
The high percentage mortality of laboratory-reared strains of Ae. aegypti and Ae. albopictus (above the WHO 80% mortality threshold) was probably stemmed from the correct formulation of deltamethrin SC-PE used on the painted cement walls. The SC formulation was found to be more suitable on cement wall surfaces than another deltamethrin formulation of WG in a recent study [23]. Furthermore, the higher mean of mortality for both Aedes spp. in the high-rise residences in comparison to the low-rise residences was most likely, in part, due to environmental influences. Since the walls in terrace houses were exposed to rainfalls and sunlight, the rate of insecticide removal from the wall surfaces of terrace houses was higher than the sheltered walls in the apartment. Besides, there appeared to be a gradual accumulation of insecticide on the wall going from spray cycle 1 to cycle 3, influencing the knockdown and mortality rates. It was observed that cycle 3 exhibited mortality of >80% until week 16 for both Aedes spp.
in the high-rise residences in comparison to cycle 1 which lasted until week 14 for Ae. aegypti and week 12 for Ae. albopictus.
The impact of vector control on reducing Aedes density is well established [45], but the evidence of the efficacy of vector control in reducing the incidence of dengue is limited. Indeed, a key limitation was that our study was not designed to detect such a relationship. To our knowledge, only a few studies have reported a positive relationship between mosquitos abundance and the number of dengue cases [46] [47]. Interestingly, Guo et al. [47] found that the increase in Ae. albopictus abundance with a lag time of one month is associated with dengue outbreak in the region. In their critical review, Wilson et al. [48] reported that shortcomings in the study design, conduct, analysis, and reporting of vector control studies are the common factors that limit the assessment of the efficacy of vector control programs on the epidemiological outcomes i.e. reduction of human cases. Recommendations on the design and conduct of vector-control studies that would provide evidence of such interventions on the epidemiology of the targeted vector-borne disease have been published recently [48] [49], which could be applied in the future studies. Thinking beyond dengue control, the recent outbreaks of Zika virus that fol-

Conclusion
In summary, we report on the initial evidence of deltamethrin SC-PE in reducing Aedes populations in the low-rise and high-rise housings. This study was useful in providing an estimation of the performance of deltamethrin SC-PE in ORS. We suggested that the ORS needs to be completed in three spray rounds of a four-month cycle to optimize its efficacy. Perhaps more well-designed randomized controlled studies with both entomological and epidemiological endpoints are warranted to better understand the true entomological effects of ORS and its impact on dengue incidence. Our results showed that ORS is a promising tool in the dengue vector control and like IRS in malaria control, it is a powerful and effective method if conducted correctly. However, similar to the IRS, several factors should be considered when conducting ORS. These include the requirement for skillful staffs to ensure uniform application of the insecticide, household's accessibility, and suitability of wall surfaces and insecticides. This study did develop excellent community trust and support that could help pave the way for further trials aiming at evaluating the impact of ORS on the incidence of vector-borne diseases such as dengue, chikungunya, and Zika.