Climatic factors and population density of Lutzomyia longipalpis (Lutz & Neiva, 1912) in an urban endemic area of visceral leishmaniasis in midwest Brazil

Authors


ABSTRACT

The life cycle of vectors and the reservoirs that participate in the chain of infectious diseases have a strong relationship with the environmental dynamics of the ecosystems in which they live. Oscillations in population abundance and seasonality of insects can be explained by factors inherent in each region and time period. Therefore, knowledge of the relationship and influence of environmental factors on the population of Lutzomyia longipalpis is necessary because of the high incidence of visceral leishmaniasis (VL) in Brazil. This study evaluates the influence of abiotic variables on the population density and seasonal behavior of L. longipalpis in an urban endemic area of VL in Brazil. The sand fly captures were performed every two months between November, 2009 and November, 2010 in the peridomicile of 13 randomly selected residences. We captured 1,367 specimens of L. longipalpis, and the ratio of male/female flies was 2.86:1. The comparison of the total male specimens in the two seasons showed a statistical difference in the wet season, but there was no significant difference when considering the total females. With respect to climatic variables, a significant negative association was observed only with wind speed. During periods of high wind speeds, the population density of this vector decreased. The presence of L. longipalpis was found in all months of the study with bimodal behavior and population peaks during the wet season.

INTRODUCTION

Vector-borne infectious diseases are an important cause of morbidity and mortality worldwide. The life cycle of vectors and the reservoirs that participate in the chain of disease transmission have a strong relationship with the environmental dynamics of the ecosystems in which they live (Barcellos et al. 2009).

Characterized as an important public health problem worldwide, visceral leishmaniasis (VL) has been a significant cause of death in Brazil, particularly after the beginning of urbanization 80 years ago, with the emergence of parasites in different urban areas in Brazil (Costa et al. 1990, 2007). In Mato Grosso do Sul, a state located in the Midwest region of Brazil, VL remains in geographical expansion affecting the rural and urban areas, and young and adults of any age group (Oliveira et al. 2006a). In 2010, 221 cases were registered in the state, and of those, 127 occurred in the city of Campo Grande (Mato Grosso do Sul 2011).

The standard of transmission and distribution of a disease vector can be modified by environmental factors and the easy adaptation of the vector associated with migration, as observed in the sand fly Lutzomyia longipalpis (Bejarano et al. 2002). In recent years, there have been several studies that correlated health to the environment by assessing the impact exerted by abiotic factors on the occurrence and abundance of arthropod vectors of diseases (Kakitani et al. 2003, Vazquez-Prokopec et al. 2006).

In Campo Grande, we observed a significant diversity of sand fly species in the urban region, with the presence of L. longipalpis, a proven vector of Leishmania infantum chagasi, Nyssomyia whitmani, Bichromomyia flaviscutellata, and Migonemyia migonei, which are vectors of cutaneous Leishmania species (Oliveira et al. 2003). Oliveira et al. 2008 showed that L. longipalpis has a significant seasonal variation in the city, predominantly in the rainy season, overlapping with higher reporting VL cases.

Oscillations in population abundance and seasonality of insects can be explained by factors inherent in each region and time period. Therefore, knowledge of the relationship and influence of environmental factors on the population of L. longipalpis is necessary because of the high mortality of VL in Brazil, especially in Campo Grande (Oliveira et al. 2010). Thus, this study evaluates the influence of abiotic variables on the population density and seasonal behavior of L. longipalpis in an urban endemic area of VL in Brazil.

MATERIALS AND METHODS

Study area

The municipality of Campo Grande (20°26′34′′S, 54°38′47′′W) has a total area of 8118.4 km2 and is located geographically in the central portion of the state, occupying 2.27% of its area total, and its altitude varies between 500 and 675 m above sea level. The city is situated in the neotropical phytogeographic areas of the Cerrado region. The climate of Campo Grande, according to the classification of Köppen, is located in the transition zone between the subtype and drought, without a mesothermal humid subtype (Aw) tropical humid, characterized by irregular distribution of rainfall annually, with the occurrence of a well-defined dry period during the colder months of the year and a rainy season during the summer months (Rohli and Veja 2008).

The study area includes a neighborhood located in the western city limits (Figure 1). This location was classified as an area of moderate transmission of VL by the State Department of Health and with high density of sand flies in accordance with an entomological survey conducted by the Center for Zoonosis Control (CCZ) of the Municipal Health.

Figure 1.

Study area.

Data collection

The sand fly captures were performed between November, 2009 and November, 2010 in the peridomicile of 13 residences around randomly selected houses. Samples were collected every two months. During the month of collection, we collected one week for a total of four samples per period. In each location, three modified light traps were installed between 17:00 and 7:00, totaling 39 traps.

Climatic data (temperature, relative humidity, rainfall, and wind speed) for the months of study were obtained from the meteorological database of the Center for Monitoring Weather, Climate, and Water of Mato Grosso do Sul that is linked to the National Institute of Meteorology (Inmet).

With the exception of rainfall, which was considered as the total value for each month, we calculated the monthly arithmetic average of the other variables mentioned. Averages were grouped for spring and summer in the wet season (September, October, November, December, January, and February) and autumn and winter in the dry season (March, April, May, June, July, and August).

The SPSS version 10.0 was used for statistical analysis. Simple linear correlation analysis (Spearman model) was used to estimate the association between the populations of males, females, and total males and females with climatic variables. Chi-square was used to compare the population density of L. longipalpis in the wet and dry seasons.

RESULTS

We captured 1,367 specimens of L. longipalpis. The ratio of male/female flies was 2.9:1. During the wet season, maximum and minimum average temperatures of 31° C and 21° C, respectively were recorded, and the humidity remained high during most of season with rainfall of 1,417 mm during the six months. In the dry season, the average maximum temperature was 29° C, and the average minimum temperature was 17° C, with total rainfall of 277 mm. The number of insects caught in the dry and wet seasons can be seen in Table 1. The comparison of the total male specimens in the two seasons showed a statistical difference in the wet season (χ2 = 55.45; GL = 1; p < 0.0001), but there was no significant difference when considering the total number of females (χ2 = 0.28; GL = 1).

Table 1. Number of sand flies in accordance with period and place of capture (PC), Campo Grande, MS, in 2009–2010.
PCWet seasonDry seasonTotal
MFMFMFMF
14609625696716192914
2945114519
31558323831
4211673281947
51221313518
6191069251944
7737614923
862228412
94311549
105113318422
1155439817
12311641307246118
1331193712683199
Total6251823881721,0133541,367

The relationship between the annual change in population density of L. longipalpis and the monthly average of climate variables is shown in Figure 2. The decrease in population density of sand flies corresponds with the fall in relative humidity and rainfall in August and September, 2010. The linear correlation analysis using the Spearman method and meteorological variables between the populations of L. longipalpis showed a negative association only with wind speed (p < 0.05); the relationship is inversely proportional (Figure 3). As wind speed increases, there is a reduction of captured specimens. Other correlations between the density of sand flies with temperature, rainfall, and relative humidity were not statistically significant.

Figure 2.

The number of sand flies captured in CDC light traps according to the climatic variables, Campo Grande, Brazil, 2009–2010 (n = 1,367).

Figure 3.

Linear correlation coefficients* between wind speed (km/h) and the density of sand flies, Campo Grande, MS, in 2009–2010. * p < 0.05.

DISCUSSION

The mode of development or growth of insects may depend on several factors including the type and amount of food available, the presence of predators, as well as abiotic factors such as humidity, temperature, and altitude, among others (Gullan and Cranston 2008). According to Deane and Deane 1955, the density of L. longipalpis is clearly influenced by seasons, with an increase in the rainy season, which favors the growth of vegetation and leads to proliferation of sand flies and other insects.

In this work, we found the presence of this vector in every month during the one year of capture, marked with a peak in January and February, 2010, which together amounted to 500 mm of precipitation and an average temperature of 25.21° C. These data are consistent with those of Oliveira et al. 2008 and Silva et al. 2007 who reported the presence of this insect monthly during two years of collection, with sharp peaks in the month of February in the urban area of Campo Grande. Galati et al. 2003 described similar results for the same species in the city of Bonito, MS, with a high peak in February followed by lower peaks at intervals of two or three months. Phlebotomus papatasi, a vector of Leishmania major in the Old World, was also present throughout the year and showed a bimodal pattern, with a record of the highest densities of population between 32 and 36° C (Boussaa et al. 2005).

This trend of dominance in the rainy season observed in this study has been reported in some parts of Brazil, as in the northeast, midwest, and southeast (Deane 1956, Galati et al. 1997, Rebêlo 2001, Oliveira et al. 2003, 2006b, Margonari et al. 2006, Resende et al. 2006). In Colombia, Morrison et al. 1995 reported the same seasonal pattern for L. longipalpis significant and a positive association between the abundance of females and relative humidity and rainfall.

A sharp drop in the population density of sand flies, coincident with the fall in relative humidity, was observed in August and September, 2010. Aguiar and Medeiros 2003 explained that relative humidity is the determining factor for the maintenance of these flies in nature since conditions of low humidity can keep them in their shelters until it rises again. We found no significant difference in the total number of females caught between both seasons. This finding has epidemiological importance since it suggests female L. longipalpis are present and can maintain disease transmission throughout the year.

The correlation analysis performed between climatic variables and population density of sand flies showed statistically significant negative correlation with wind speed. This inverse relationship can be evidenced clearly between July and September, 2010. At the time, we recorded higher values for wind speed and smaller amounts of captured specimens. Some studies corroborate these observations and relate the decrease in the number of insects captured by the high wind speed (Zeledón et al. 1984, Ximenes et al. 2006).

By analyzing the influence of wind on the biting activity and population frequency of Anopheles marajoara (Diptera: Culicidae), Kakitani et al. 2003 found highly significant values, and on wind speed less than 3.0 km/h, the frequency of mosquitoes decreased considerably. Besides the negative association with wind speed, Ximenes et al. 2006 reported a positive association with temperature for the total females, with relative humidity for a total of males and females, and with total rainfall for females of L. longipalpis in the forest region of northeast Brazil.

In Brazil, wind speed is higher in the driest months of the year and it is less humid, especially during the months of June, July, and August. So, the other climate variables, especially rainfall and relative humidity, can influence of the behavior of these insects. Some general characteristics, such as small size, which can vary from 2 to 4 mm, and thickness of thin chitinous cuticle that covers the body surface make these flies vulnerable and exposed to bad weather such as an increased wind speed. Therefore, at the time of these events, sand flies tend to stay in their shelters where the elements can offer a favorable microclimate that protects them from desiccation (Forattini 1973).

The other correlations between climatic variables (temperature, relative humidity, and rainfall) and the data did not show any significant aspects for sand flies. Although there was no statistical difference in the analysis mentioned above, it became evident that there is a tendency for species to be more abundant in the wet season. These findings corroborate those reported by Rebêlo 2001 and Resende et al. 2006, which described increases in the number of sand flies captured between October and March in São Luis and Belo Horizonte, respectively.

An important issue that deserves attention and has been little discussed in studies of seasonality is the climate change that has occurred in recent years, mainly in urban areas. In Brazil, some studies indicate that the semi-arid region of the northeastern, northern, and eastern Amazon and the southern neighborhoods are markedly affected by El Nino. In the south, there is an increase in precipitation, particularly during the spring of the first year and in late autumn and early winter of the second year (Barcellos et al. 2009, Sampaio 2000). Due to these changes and the easy adaptation of the vector, it has been observed that expansion of areas with autochthonous VL and reported the presence of L. longipalpis in regions with subtropical climate, as the state of Rio Grande do Sul (Souza et al. 2009, Werneck 2010).

In conclusion, we noticed the presence of L. longipalpis in all months of the study with bimodal behavior and population peaks during the wet season. The negative correlation between wind speed and the presence of these flies was statistically significant, so it was evident that during periods of high wind speeds, the population density of L. longipalpis decreased.

Acknowledgments

We thank the Center of Zoonosis Control (CCZ) of the Municipal Health Secretariat of Campo Grande for providing the necessary logistic support and helping us in field activities. We are also grateful to the residents of houses who allowed us to make captures at their residences, and to Mrs. Aline Casaril for helping us with some of the experiments.

Ancillary