Portal de Periódicos da CAPES

Personal protection by long-lasting insecticidal hammocks against the bites of forest malaria vectors

2010; Wiley; Volume: 15; Issue: 3 Linguagem: Inglês

10.1111/j.1365-3156.2009.02457.x

1365-3156

Tho Sochantha, Wim Van Bortel, S. Savonnaroth, Tanguy Marcotty, Niko Speybroeck, Marc Coosemans,

Insect Pest Control Strategies

Objective In Southeast Asia, malaria vectors bite outside the houses before bedtime, and forest dwellers rarely use insecticide-treated nets (ITNs). Thus, we tested the protection of long-lasting insecticidal hammocks (LLIH) using Olyset™ technology against exophagic vectors in two forest villages of Cambodia. Methods In each village, we conducted two entomological surveys (middle and end of the rainy season), each lasting 10 consecutive nights. These comprised human landing collections during the whole night by people sitting outside in the hammocks. Five households were selected per village, and for each household, two fixed positions were allocated: one for the control and one for the treated hammock. Results In total, 6449 mosquitoes were collected from control hammocks compared to 4481 in treated hammocks. Personal protection conferred by the hammocks was 46% (CI 95%: 35–55%) against the bites of Anopheles minimus. A significant reduction of An. dirus bites (46% CI 95%: 25–62%) was only observed at the end of the rainy season. An. maculatus and culicines bites were only reduced in one of the two study sites. Conclusion Even if this LLIH is not inducing full protection against the bites of malaria vectors, it could prove effective in protecting forest workers and villagers before sleeping time. LLIH can be an additional and valuable tool in eliminating artemisinin-resistant malaria in the region. Protection individuelle par des hamacs à insecticide durable contre les piqûres des vecteurs de la malaria en zone forestière Objectif: En Asie du Sud, les vecteurs de la malaria piquent à l'extérieur des maisons avant le coucher et les habitants des forêts utilisent rarement des moustiquaires imprégnées d'insecticide (MII). Nous avons alors testé la protection conférée par des hamacs imprégnés d'insecticide durable (HII) en utilisant la technologie Olyset™ contre les vecteurs exophages dans deux villages de forêt au Cambodge. Méthodes: Dans chaque village, nous avons mené deux enquêtes entomologiques (au milieu et à la fin de la saison des pluies), chacune durant 10 nuits consécutives. Cela comprenait la collecte des vecteurs atterrissant sur les gens tout au long de la nuit et sur les personnes assises à l'extérieur dans les hamacs. Cinq ménages ont été sélectionnés par village et pour chaque ménage deux postes fixes ont été définis: un servant de contrôle et l'autre pour le hamac traité. Résultats: Au total, 6449 moustiques ont été recueillis dans les hamacs contrôle par rapport à 4481 dans les hamacs traités. La protection individuelle conférée par les hamacs était de 46% (IC95%: 35-55%) contre les piqûres d'Anopheles minimus. Une réduction significative des piqûres d'An. dirus (46%; IC95%: 25-62%) a seulement été observée à la fin de la saison des pluies. Les piqûres d'An. maculatus et culicinés étaient seulement réduites dans l'un des deux sites d'étude. Conclusion: Même si ces hamacs traités n'induisent pas une protection totale contre les piqûres des vecteurs de la malaria, ils pourraient être efficaces dans la protection des travailleurs forestiers et des villageois avant l'heure du coucher. Les hamacs traités peuvent être un outil supplémentaire et précieux dans l'élimination de la malaria résistante à l'artémisinine dans la région. Protección personal mediante el uso de hamacas impregnadas con insecticida de larga duración contra las mordeduras de vectores de malaria forestales Objetivo: En el sudeste asiático, los vectores de malaria pican fuera de las casas antes de la hora de dormir, y los habitantes de los bosques rara vez utilizan mosquiteras impregnadas de insecticida (MIIs). Por lo tanto hemos probado la protección que resulta del uso de hamacas impregnadas con insecticida de larga duración (HILD) con la tecnología Olyset™ frente a vectores esofágicos en dos poblados en bosques de Cambodia. Métodos: En cada poblado, se realizaron dos estudios entomológicos (a la mitad y al final de la época de lluvias), cada uno con una duración de 10 noches consecutivas. Los estudios incluían colectas con cebo humano, realizadas durante toda la noche por personas que estaban sentadas fuera en las hamacas. Se seleccionaron cinco hogares por poblado y en cada hogar se seleccionaron dos posiciones fijas: una para el control y una para la hamaca tratada con insecticida. Resultados: En total se recolectaron 6449 mosquitos de las hamacas controles, comparadas con 4481 en las hamacas tratadas. La protección personal conferida por las hamacas era del 46% (IC 95%: 35-55%) frente a las mordeduras de Anopheles minimus. Se observó una reducción significativa en las mordeduras de An. dirus (46% IC 95%: 25-62%) solo al final de la estación de lluvias. Las mordeduras de An. maculatus y culicines solo se redujeron en uno de los dos lugares de estudio. Conclusión: Aún si las HILD no brindan una protección completa frente a las mordeduras de vectores de malaria, podrían ser efectivas para proteger a los trabajadores y habitantes de los bosques antes de la hora de dormir. Las HILD podrían ser una herramienta adicional y valiosa a la hora de eliminar la malaria resistente a artemisinina presente en la región. In Southeast Asia, forest malaria remains a big challenge for malaria control programmes. The major vectors, Anopheles dirus, Anopheles minimus and Anopheles maculates, are mainly biting outside, particularly when houses have fully closed walls and are built on stilts. Except in the northern part of Vietnam, almost no vectors rest inside the house after biting (Trung et al. 2005). This outdoor behaviour may explain the persistence of malaria transmission after indoor spraying with insecticides (Coosemans & Van Bortel 2007). Moreover, a significant proportion of malaria vectors will bite before bedtime, which reduces the efficacy of insecticide-treated nets. This is particularly true for An. maculatus and to some extent for An. dirus according to the locality (Trung et al. 2005). Community-based studies in Central Vietnam showed that regular forest activity was a very strong risk factor for malaria infection (Erhart et al. 2005; Ngo et al. 2008). This means that traditional vector control methods are not fully effective in controlling forest malaria. Indoor spraying with residual insecticides will not be effective against the malaria vectors because of their exophilic behaviour and insecticide-treated nets are not used by forest workers during the night. In the Amazon region, insecticide-treated hammock nets were effective in reducing malaria incidence (Magris et al. 2007). These insecticide-treated nets are hung over the hammocks only at bedtime; the main vector An. darlingi is mainly active between midnight and 4:00 h (Rubio-Palis 1995). Malaria vectors in Southeast Asia bite outside in the early evening before people go to sleep (Trung et al. 2005; Coosemans & Van Bortel 2007). However, most villagers and forest workers rest in hammocks in the early evening. Insecticide-treated hammocks can, therefore, be an alternative to prevent the bites of exophagic and early biting malaria vectors. Additional and adapted vector control tools are more than welcome, particularly because artemisinin-resistant Plasmodium falciparum is spreading at the border of Cambodia and Thailand (WHO 2008). In a forested area of central Vietnam, long-lasting insecticidal hammocks (LLIH) proved very effective in reducing malaria incidence and prevalence (Thang et al. 2009). To better understand the entomological impact, we assessed the personal protection of LLIH against the bites of malaria vector species and culicines in villages surrounded by forest in the western part of Cambodia. The study took place in two villages of western Cambodia: one in the province of Pailin (village of Pang Rolim, 12.788°N – 102.691°E) where An.minimus is the dominant vector, and one in the province of Pursat (village of Dey Krahorm Leu, 12.274°N – 102.952°E), where An. dirus predominates. No insecticide resistance against pyrethroids was observed in these sites (Van Bortel et al. 2008). The study village in Pailin is in a highly deforested area, whereas the village in Pursat is in an evergreen forested environment. Wooden houses are often built on stilts covered with thatch or iron roofs. Before bedtime (22 h) villagers spend the evening outside or under their house in hammocks. More than 90% of the inhabitants sleep under a non-treated bed net inside the house. Some people have regular forest activities such as farming, logging or hunting. The proportion of villagers sleeping in the forest is higher in Pursat (about 30%) than in Pailin (20%). The same design of LLIH tested during the community–based intervention trial carried out in central Vietnam (Thang et al. 2009) was used in this study. Briefly, hammocks (2.07 m × 1.03 m) were made of tick green (kaki) cotton fabric. On each end, a woven rope (1.5 cm × 3.70 m) was attached for hanging it. A layer of Olyset netting material (100% high-density polyethylene blended with permethrin 2%, with a mesh of 4 mm × 4 mm – Sumitomo Chemical Co. Tokyo, Japan) was sewed on the backside of the hammock. Along one of the long sides, a flap of Olyset (0.4 m large) was sewed, with which the person in the hammock can cover himself or herself (Figure 1). Control hammocks had the same design, but the netting material was not treated. Design and use of the long-lasting insecticidal hammock. At the edge of each village, five households were selected, with a minimal distance of about 200 m between the households. For each selected household, two fixed sites at a distance of about 10 m were chosen outdoors. Each site was allocated to a control or a treated hammock. Hammocks were coded and labelled at central level before the study started, and their treatment status was not known by the supervisors in the field and the collectors. Hammocks were stored in the corresponding labelled plastic bags before and after use. Collectors were only allowed to touch their own hammock. Two surveys of 10 nights were performed in each village (Pailin: July and October 2006; Pursat: August and November 2006). Outdoor human landing collections of mosquitoes lasted each night from 18:00 to 06:00 h whilst collectors were sitting in the hammock at the side of the flap. In each village, collectors were organized in 10 teams of three persons recruited amongst the village community, and rotation of collectors within a team was performed every 4 h. In each hammock, one person collected the mosquitoes landing (before biting) on his bare legs using a torch and collector tube. The specimens were transferred individually in tubes and collected every hour by a supervisor. Teams rotated after each night according to a Latin Square Design, so that none of them occupied the same position during one survey. Anopheline mosquitoes were separated from culicines and identified morphologically using an illustrated key for anophelines of Southeast Asia (adapted from NIMPE 1987). The collected culicines were not identified to species level but were considered in the analysis to measure the protection of treated hammocks against mosquito nuisance. Molecular identifications of An. dirus sensu lato (Walton et al. 1999) and An. minimus s.l. (Van Bortel et al. 2000) have been reported in previous studies (Van Bortel et al. 2008; Obsomer et al. 2007, W. Van Bortel, personal communication). In both study sites, only An. dirus sensu stricto and An. minimus s.s. were observed. For An. maculatus s.l., no further identification was performed. Data were entered twice in Microsoft Access and checked for accuracy. For each species, analyses were made for all specimens collected during the whole night (18:00–6:00) and another for the ones collected before 22:00 h. Data were analysed in STATA 9 (Stata Corporation, College Station) using a robust negative binomial regression model. Hammock pairs were used as primary sampling units. Explanatory variables were treatment status of hammocks, village, survey and the interaction between them. The models were simplified by removing non-significant explanatory variables (P > 0.1) and assessing the reduced models using the adjusted Wald test (P > 0.05). The national ethics review board for health research of Cambodia and the Institutional Review Board of the Institute of Tropical Medicine of Antwerp approved the protocol of the study. Collectors and householders were informed in Khmer of the purpose of the study, and an informed consent was signed by each collector. Participation was voluntary, and they were allowed to withdraw from the study at any time. All collectors were checked for malaria parasites 12 days after the last collection of each of the two surveys. A full course treatment, according to the national guidelines, was provided by the team leader (Medical Doctor) of the National Malaria Control Programme when collectors tested positive. Access to malaria diagnosis and treatment was guaranteed throughout the study. Of 11 055 mosquitoes collected, 62% were anopheline mosquitoes. Anophelines were more abundant in Pailin (63%) than in Pursat (37%). In Pailin, An. minimus was the predominant anopheline species (67.2%), followed by An. maculatus (20.6%) and An. dirus (9.9%). In Pursat, An. dirus predominates (52.0%), followed by An. maculatus (30.7%) and An. minimus (12.3%). The proportions of bites before 22 h were 43% (n = 4197) by culicinines, 56% (n = 1670) by An. maculatus, 36% (n = 1751) by An. dirus and 28% (n = 3209) by An. minimus. The hourly biting rhythm of the main malaria vectors is given in Figure 2 for mosquitoes collected in the control hammocks, and Figure 3 provides the estimated number of bites/man/night for the different species in different settings. The treated hammocks had a significant impact on An. minimus bites (reduction by about 45%) in both villages and for both surveys and for mosquitoes collected before 22.00 h (P = 0.004) or during the entire night (P < 0.001) (Table 1). Hourly biting rhythm of the three vector species (only for control hammocks). Average number of bites per man per night of the collected mosquitoes (confidence intervals 95%) with long-lasting insecticidal hammocks (dark) compared to control ones (grey) in two villages during two surveys (Survey 1: July–August; Survey 2: October–November 2006). For An. dirus collected during the whole night, a significant interaction was observed between survey and treatment (P = 0.036), and treatment had only a significant effect during the second survey (reduction of 46%–P = 0.005). The interaction treatment–village survey for An. dirus collected before 22.00 h was significant (P = 0.004) and did not allow further simplification of the model. For An. maculatus, the interaction of treatment–village was just above the significance threshold (P = 0.068) for mosquitoes collected during the whole night but was significant for collections made before 22.00 h (P = 0.033). In fact, the protective effect of treated hammocks against this species was only significant in Pailin both for the whole night (50% reduction –P = 0.002) and before 22.00 h (56% reduction –P = 0.002). For culicines, too, the effect was significant only in Pailin (reduction of 29%–P = 0.023 and 37%–P = 0.002 for the whole night or before 22.00 h, respectively). The early biting behaviour of An. maculatus and to a lesser extent of An. dirus and An. minimus, confirms previous observations (Trung et al. 2005). The personal protective effect of LLIHs against exophagic vectors and nuisance mosquitoes was variable according to species, villages and surveys. About half of the An. minimus bites can be avoided by just sitting in the LLIH. A similar result was obtained against An. dirus but only during the second survey (end of the rainy season), and no evidence of protection was found in the middle of the rainy season. A significant protection against An. maculatus and culicines was only obtained in one village (Pailin). There is no sound explanation for this seasonal and geographical variation in efficacy. Molecular identification in the two villages confirmed the presence of only An. dirus s.s. and An. minimus s.s. (Van Bortel et al. 2008; Obsomer et al. 2007; W. Van Bortel, personal communication), so that no other sibling species are involved to explain this disparity in efficacy. However, different behavioural patterns of the same vector have been observed in the region (Trung et al. 2005), and this may interfere with the response to the treated material according to place and season. Behavioural actions of insecticide-treated materials in disrupting man–vector contact are more complex than initially thought (Grieco et al. 2007), and further studies are needed to understand the influence of the environmental conditions on the performance of these tools. With a similar design of hammocks, Hougard et al. (2007) obtained a personal protection of 95% against An. gambiae during an experimental hut trial. This high level of protection can be explained by the fact that people slept indoors in the treated hammocks and were covered with a net containing permethrin. In our trial, personal protection has probably been underestimated as volunteers just sat in the hammocks the whole night. This sitting position to collect the mosquitoes was chosen to reproduce the habit of people spending the evening outside. The purpose of the treated flap sewed along one side of the hammock is to reduce the landing of mosquitoes on the legs. This is particularly effective when the flaps are treated with permethrin, a repellent and irritant to mosquitoes. Considering the limited protective efficacy of insecticide-treated nets in Southeast Asia (Sochantha et al. 2006), LLIHs should be regarded as an additional tool in the control of malaria. These LLIHs were proved to be very effective in preventing a substantial number of clinical cases and malaria infections in high endemic forest villages of Vietnam. These encouraging results were obtained despite the low rate of use amongst forest workers, but the use of it in the villages was sufficient to obtain a good impact (Thang et al. 2009). A further sensitization of forest workers is certainly needed. Even if these LLIHs are not inducing full protection, they could prove effective in protecting forest workers and villagers during the evening before bedtime. Besides scaling up artemisinin combination treatment and long-lasting insecticidal nets, LLIHs can help to control the spread of artemisinin-resistant falciparum malaria in Cambodia (Maude et al. 2009). We are grateful to the manufacturer Sumitomo Chemical Co., Inc. (Tokyo, Japan) for providing the experimental LLIH. This study was conducted within the framework of the Institutional collaboration between CNM-Cambodia and the Institute of Tropical Medicine of Antwerp Belgium, supported by the Belgian Co-operation.