Abstract
The oviposition level between ovitraps placed in human dwellings and on public paved areas for the monitoring of Aedes aegypti was compared, and proposes the best ovitrap installation sites on public paved areas in a neighborhood of Buenos Aires City, Argentina. A total of 60 randomly distributed ovitraps were placed in human dwellings (n=38) and public paved areas (n=22), and examined weekly over a ten-week period. The Ovitrap Positivity Index (OPI) and the Egg Density Index (EDI) were used as indicators of oviposition level. The indexes were calculated on a weekly basis. Environmental variables evaluated for public paved areas were type of corridor, tree height and dwelling height. OPI and EDI values were higher for dwellings than for public paved areas. There was a positive correlation between the infestation indices in dwellings and in public paved areas. In the latter there was a positive association between low OPI and EDI values and tall trees, avenue as type corridor, and dwellings with two stories or more, whereas there was a positive association between high values of these indices and short and medium trees, and a negative association with tall trees, avenue, and dwellings with two stories or more. Our results suggest that the sensitivity of ovitraps in monitoring of A. aegypti in public paved areas is lower than in dwellings, and this sensitivity would increase if considering some environmental variables.
Introduction
Aedes aegypti (Linneo, 1752) is the main vector of dengue virus in America. Although this mosquito was considered eradicated from Argentina in 1963 (Carcavallo 1968), in 1986 A. aegypti reinfestation was detected (Curto et al. 2002), and since then been reported several dengue fever outbreaks in this country (Vezzani and Carbajo 2008). Aedes aegypti was found in urban areas adjacent to Buenos Aires City in 1991 (Campos et al. 1993) and within the city in 1995 (Junin et al. 1995). The introduction of the virus in Buenos Aires was detected in 2007 (MSN 2007), reported in early 2009 an outbreak of indigenous dengue in the Buenos Aires Metropolitan area (Seijo et al. 2009). This fact, together with high infestation levels of vector (Schweigmann et al. 2002; Carbajo et al. 2004) may increase the risk of virus transmission in this city. The distribution of A. aegypti in Buenos Aires shows a spatio-temporal pattern (Carbajo et al. 2004). Peaks of abundance have been reported in February, March and April and oviposition activity from October to May (Schweigmann et al. 2002). The highest oviposition activity has been registered in neighbourhoods with low-rise houses and a few multi-story buildings, which are located in the periphery of the city (Carbajo et al. 2004).
The use of an efficient tool for the detection of vectors allows a better knowledge of the actual vectorial status. In particular, the oviposition trap (ovitrap) is regarded as one of the most sensitive and cheap tools for surveillance of A. aegypti, especially in situations of low vector density (Jakob and Bevier 1969a; Chadee 1991; PAHO 1994; Focks 2003). Despite the wide use of ovitraps (PAHO 1994), studies on factors affecting their efficiency are scarce. Some criteria have been established for the location of ovitraps based on environmental features (Fay and Eliason 1966; Jakob and Bevier 1969b). More recently, some authors have investigated the relationship between the degree of detection and ovitrap installation sites in residences (Chadee 1991, 1992; Dibo et al. 2005).
Generally, it is accepted that the breeding sites of A. aegypti are placed in the dwellings range (PAHO 1994), reason why in many cities the ovitraps are placed in the dwellings (e.g., Chadee 1991, 1992; Dibo et al. 2005; Lenhart et al. 2005). However, the use of ovitraps for monitoring is hampered if the permission to enter a house is refused (Calvo 2008; Stein and Oria 2002), and this may lead to the failure of control measures against A. aegypti (Chadee 1988). Besides dwellings, these devices have been laid on public paved areas to investigate the spatial distribution of the vector (Chadee 1990; Ai-lenn and Song 2000; Schweigmann et al. 2002; Carbajo et al. 2004, 2006). However, no research has so far been conducted to evaluate the use of ovitraps in this environment. Therefore, we present the results of a preliminary study whose the objectives were to compare oviposition levels between ovitraps placed in dwellings and on public paved areas for the surveillance of A. aegypti, and to identify the best ovitrap installation sites on public paved areas.
Materials and methods
Study area
The climate of Buenos Aires City (34°35'S 58°29'W) is influenced by the De la Plata River. It is temperate and shows a marked seasonality, with an annual mean precipitation of 1,076 mm and a mean temperature of 17.4 °C (National Weather Service).
The study was performed in the periphery of Buenos Aires City, in the neighbourhood of Villa Pueyrredón, in the northeast of this city. An area of approximately four hectares of this neighborhood was chosen by the project "Abordaje ecosistémico para la prevención y el control del vector del dengue en Uruguay y Argentina" by support of IDRC. Our study was carried out in this area. The study area had a total of 328 buildings: residential a 97.6% (320 houses/328 buildings), commercial a 2.1% (7/328), and only one property empty. No park or square was included in the study area.
Villa Pueyrredón is characterised by one-story houses with large gardens and yards, blocks showing a high proportion of vegetation cover and many trees on the pavements. The population in the neighbourhood is 40,235 inhabitants (INDEC 2001). In Villa Pueyrredón, A. aegypti infestation level showed a House Index (the percentage of houses infested with larvae and/or pupae) and a Breteau Index (the number of positive containers divided by number of houses examined per 100) (Breteau 1954) of 50 and 85.7, respectively, for the period February-April 2006.
Oviposition activity of A. aegypti was monitored weekly with ovitraps during ten consecutive weeks between February and April 2006, no were made vector control interventions with insecticide (chemical or biological) during this period. The ovitraps were set at ground level, in shaded or partially shaded sites close to vegetation. Each ovitrap consisted of a black 330 ml-glass jar containing about 100 ml of tap water and a hardboard paddle (10 x 3.2 x 0.35 cm); it was held in vertical position by a paper clip, with its rough side facing the centre. This wood-based material provides a surface suitable for A. aegypti females to lay eggs (Service 1976). No infusion was added to the traps.
Comparison of oviposition levels between public paved areas and dwellings
In an area of approximately 4 ha a total of 60 ovitraps were randomly distributed, 38 ovitraps were placed in dwellings and 22 at the base of trees on public paved arcas. In cach dwelling (house) was placed one ovitrap in yard or garden. None of the ovitraps placed in public paved areas was installed immediately in front of a house that had an ovitrap inside. Mean distance between ovitraps was 25 ± 12.1 m (mean ± SD). In ovitraps placed on public paved areas, the number of traps lost ranged between 0-3/week. Each week, the hardboard paddles and the water in the ovitraps were replaced, and the jars washed. The number of eggs on each hardboard paddle was counted under stereoscopic microscope. Ovitraps were considered positive (infested by A. aegypti) when at least one egg was detected. Each week four paddles randomly selected were conditioned for eggs hatching and larvae rearing. Fourth instar larvae were identified with taxonomic key (Darsie 1985).
The Ovitrap Positivity Index (OPI) and the Egg Density Index (EDI) were used as indicators of oviposition level. These indexes were calculated on a weekly basis, with OPI= (nº positive ovitraps/total nº examined ovitraps) x 100, and EDI= total nº eggs/total nº positive ovitraps (Gomes 1998).
The Mann-Whitney U test (Zar 1996) was used to compare weekly values of OPI and EDI between dwellings and public paved areas. The Spearman's non-parametric correlation test (Zar 1996) was used to assess the correlation between OPI and EDI for dwellings and public paved areas.
Environmental variables of public paved areas
The three following environmental variables, located within a radius of 6 m from the ovitrap, were registered for each installation site: a) type of corridor: Street (narrow corridor with scarce vehicles and one-way traffic) or Avenue (wide corridor with heavy two-way traffic); b) tree height: Tall (taller than 6 m; smallest distance between the trunk base and edge of foliage higher than 4 m), Medium (height between 4 and 6 m; smallest distance between the trunk base and foliage edge between 2 to 4 m) or Short (height less than 4 m; smallest distance between the trunk base and edge of foliage less than 2 m); c) height of dwellings: H1 (<2 stories) and H2 (≥ 2 stories). The relationships between OPI and EDI and each of the variables were studied with correspondence analysis (Gauch 1982), dividing the values of the indices into intervals (OPI: low 0-30%, medium >30-70%, and high >70%; EDI: low 0-10 eggs, medium >10-20 eggs, and high >20 eggs).
The temperature and precipitation data used were obtained from the meteorological station at the Jorge Newbery Airport of Buenos Aires City (NCDC 2006). This meteorological station is located about 8 km of study area. The analysis was based on weekly-accumulated rainfall and mean temperatures, which were calculated from mean daily temperatures.
Results and discussion
Comparison of oviposition levels between public paved areas and dwellings
The studied period was characterised by a mean weekly temperature of 20 °C (17.4-21.7 °C) and a weekly mean accumulated rainfall of 27.3 mm (Fig. 1).
Of the 330 ovitraps data of dwellings, 65.4% were positive with a total of 10,550 eggs, whereas 50.5% of the 206 ovitraps data of public paved areas were positive with a total of 3,368 eggs.