Implication of agricultural practices in the micro-geographic heterogeneity of malaria transmission in Bouna, Côte d’Ivoire

doi:10.21203/rs.3.rs-3120408/v1

Download PDF

Research Article

Implication of agricultural practices in the micro-geographic heterogeneity of malaria transmission in Bouna, Côte d’Ivoire

https://doi.org/10.21203/rs.3.rs-3120408/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Wetlands and irrigated agricultural crops create potential breeding sites for Anopheles mosquitoes, leading to a heterogeneity in malaria transmission. In agricultural areas, heterogeneity of malaria transmission is often associated with the presence of hotspots consisting of localized clusters of higher transmission intensity. This study aims to identify micro-geographic hotspots of malaria transmission in an agricultural setting using a multidisciplinary approach.

Methods

We conducted two cross-sectional surveys at the end of the dry season and at the peak of the rainy season in rural and urban sites in Bouna, northeastern Côte d'Ivoire. We randomly selected 296 individuals from 148 farming households and collected sociological, geographical, entomological, and clinical data as well as blood samples during each visit. We compared parasitological data and Anopheles exposure (measured using entomological and immunological methods) with demographic, agricultural, and geographic data to identify drivers of malaria transmission. Heat maps combining these data were used to identify households with ongoing malaria transmission throughout the year.

Results

In rural areas, Plasmodium prevalence was consistent between the dry and the rainy seasons, with roughly half of the population presented with an infection. In urban areas, malaria transmission indicators were lower, with a parasite prevalence of less than 20%, which remained comparable between the dry and the rainy season. The presence of irrigated crops and proximity to wetlands were associated with increased Anopheles exposure. By mapping Plasmodium infection and Anopheles exposure, we identified 2 different types of hotspots of malaria transmission: micro-geographical scale and local scale hotspots.

Conclusions

The presence of wetlands in urban areas and irrigated agriculture in rural areas resulted in heterogeneity in malaria transmission on a micro-geographical scale. These specific households present particular risk of malaria transmission and could fuel malaria transmission in surrounding households. The identification of micro-geographical areas using heat maps combining several epidemiological parameters can help to identify hotspots of malaria transmission. The implementation of malaria control measures, such as seasonal chemoprophylaxis or vector control, in these areas could help to reduce the incidence of malaria and facilitate its elimination.

malaria

hotspot

microgeographic heterogeneity

agriculture practices

Substantial reductions in the global burden of malaria were noted during the past two decades in response to improvement of control measures with the mass distribution of long-lasting insecticide-treated bednets (LLIN), access to diagnosis with rapid diagnostic tests (RDT) and prompt treatment with artemisinin-based combination therapy (ACT). But since few years, progress has stagnated with global malaria incidence and deaths remain stable with over 200 million malaria cases and more than 600,000 deaths in 2021 [1], strengthening the need of new approaches to reach elimination goal. Since the COVID pandemic, we noted upsurges in many Sub-Saharan countries including Côte d’Ivoire where Plasmodium falciparum (P. falciparum) malaria is the leading cause of mortality among children [2].

P. falciparum transmission is driven by ecological factors including those that favored Anopheles breeding sites and thus Anopheles populations. The presence of wetland as well the type of agricultural crops and practices, particularly irrigated cultures can create suitable breeding habitats for Anopheles mosquitoes, because mosquitoes require stagnant water for their larval habitat. A study in Kenya showed that irrigated agriculture was associated with a higher malaria transmission rate, likely due to increased mosquito breeding in the irrigation canals [3]. Similarly, a study in Mali reported that agricultural practices, particularly irrigation, were a major risk factor for malaria transmission in the region [4]. The watering of the agricultural areas during the dry season can also promote local malaria transmission resulting in a risk of malaria transmission throughout the year [5, 6]. Thus, the presence of wetlands nearby households might also be an ecological factor that influence spatial and temporal malaria transmission dynamic resulting in a heterogeneity of transmission. The households located nearby wetlands or irrigated crops might be particularly at risk and might clustered and represented hotspots where malaria is sustained throughout the year. During the dry season, the hotspot maintains transmission within the geographic area while during the rainy season, it fuels the transmission in surrounding areas [7]. As a result, understanding the link between land use, crop practices, and malaria transmission is crucial for designing effective control strategies and reducing the disease burden in endemic regions. Although complex, identifying malaria transmission hotspots is important for malaria elimination efforts. Thus, the identification of malaria hotspots that are stable in space and time might help to implemented targeted interventions to these specific areas, such as mass drug administration (MDA) or seasonal malaria chemoprevention (SMC).

In recent years, serological marker have been developed to assess individual level of exposure of human population to Anopheles bites. More specifically, the IgG response to the Anopheles gambiae gSG6-P1 salivary peptide has been identified as a proxy of exposure to Anopheles bites [8]. Studies in malaria endemic countries have shown that the use of this serological indicator is suitable to identify heterogeneity of exposure to Anopheles bites [9, 10], to evaluate the efficacy of vector control interventions and to identifying individuals at higher risk of malaria [11, 12]

In the present study, we aimed to assess the role of wetlands and agricultural practices in malaria transmission in agricultural households in rural and urban settings in Bouna, Côte d’Ivoire using several epidemiological indicators of malaria transmission. Mapping together these epidemiological parameters helped us to identify households the most at risk of malaria transmission in the different settings.

Study sites

The study was conducted in Bouna located in the North-East of Côte d'Ivoire (9°16' N,3°00' W) at the northern border to Burkina Faso and Ghana. The climate is tropical with two seasons including a dry season which runs from November through May and a rainy season from June to October. The annual rainfall recorded in 2020 was estimated at 847.4 mm with an average temperature around 26.7°C. Bouna has approximately 193,698 inhabitants, including 47,892 households [13]. The study was carried out in 6 sites: 3 villages in rural setting (Assoum1, Téguiduo and Bromakoté) and 3 neighborhoods in urban setting (Doropo2, Dassikeledougou and Gborotchara) (Fig. 1).

Excepted Gborotchara and Bromakoté, all these study sites are characterized by the presence of large areas of wetlands, some of which are used for irrigated rice cultivation and market gardening. The presence of domestic animals such as cats and dogs are quite remarkable as well as the breeding practices of cattle, goats, and poultry. The duration of the malaria transmission season is perennial with a peak in the rainy season. The most important malaria vector in this area is An. gambiae s.l. [14].

Study Design, Procedure, and Sample Collection

Two cross-sectional surveys were carried out at the end of dry season (Mai 2021) and at the peak of the rainy season (August 2021), respectively. One hundred and forty-eight (148) agricultural worker households were randomly selected from the 2020 malaria registry from the city's general hospital and one child and one adult from each household were recruited, resulting in a total of two hundred and ninety-six (296) participants.

Each household received a long-lasting insecticide-treated mosquito nets (LLINs) at the time of inclusion in late May 2021 during LLINs mass campaign distribution by the national malaria control program (NMCP). At each visit, sociological, geographical, entomological, and clinical data were collected. Participant’s axillary temperature was measured, and thick films and blood smears were performed to determine parasite density and Plasmodium species. Thick blood smears were fixed and stained with 10% Giemsa and read at double blind by certified microscopists. Asexual parasite densities were counted against 200 microscope fields white blood cells assuming 8,000 white blood cells per microliter. A blood smear was considered negative if no parasites were observed. For immunological assays, dried blood spots (DBS) were collected at the fingertips on Whatman 3 MM filter, air-dried and stored in plastic bags at + 4°C until immunological analysis.

Mosquito collection

Adult mosquitoes were collected by human landing catches (HLC) in June and August in two reference households in each of the 6 sites. In each reference household, two catching points, one indoor and one outdoor were used to collect mosquitoes by landing catches on adult volunteers from 6.00 pm. to 6.00 am. Adult mosquito catchers gave prior informed consent, were vaccinated against yellow fever and were treated for malaria according to the national treatment guidelines. The mosquitoes were collected, and carefully stored in closed hemolysis tubes with cotton and kept in plastic bag by time slot and by catching point. The human biting rate (HBR) of An. gambiae was calculated as the average number of mosquitoes collected per person per night (bites/human/night, BHN). The mosquitoes from HLC methods were identified morphologically at the species level. This identification was made according to the key to determining the Culicinae of Mattingly and that of the Anophelinae of Gillies and of Meillon [15] .

Evaluation of the use of LLINs and inventory of agricultural practices

The use and physical integrity of the LLINs was recorded by the administration of questionnaires in all study households. Community health workers verified bednet installation after distribution of the news LLINs during the mass campaign. The questionnaires also recorded data on agricultural practices: nature of culture and type of irrigation practiced by household members.

Measurement of human IgG level to Anopheles gSG6-P1 salivary peptide.

Human IgG level against the gSG6-P1 salivary antigen of An. gambiae was measured by enzyme-linked immunosorbent assay (ELISA). Standardized dried blood spots (diameter: 1 cm) of each participant at two time points were eluted each in 300 µL phosphate buffered saline containing 0.1% Tween 20 (0.1% PBST) at 4°C for 24 hours. Subsequently, 96-well Maxisorp micro-assay plates (Nunc, Roskilde, Denmark) were coated with 100 µl of the gSG6-P1 salivary peptide (20µg/mL in PBS) for 2 hours and 30 minutes at 37°C. Plates were blocked for 1h with 200 µL of protein-free blocking buffer, pH7.4 (Thermo scientific, Rockford, USA). After washing, individual eluate was incubated in duplicate wells at 4°C overnight wells at 1/40 dilution in 0.1% PBST. Mouse biotinylated anti-human IgG antibody (BD Pharmingen, San Diego, CA) was added at a 1:2000 dilution in 1%-PBST (1 hour and 30 minutes at 37°C), and peroxidase-conjugated streptavidin (Amersham, Les Ulis, France) was then added (1/15,000; 1 hour at 37°C). Colorimetric development was carried out using ABTS (2.2′-azino-bis (3 ethylbenzthiazoline 6-sulfonic acid) diammonium; Roche Diagnostics, Lyon, France) in 50 mM citrate buffer (pH 4) containing 4% H₂O₂ and the Optical Density (OD) was measured 2h later at 405 nm. Individual results (ΔOD) are expressed as: ΔOD = ODx-ODn where ODx represents the mean of individual OD value in both wells containing the salivary antigen and ODn the individual OD value for each eluate without gSG6-P1 antigen.

Maps

The geographic coordinates of the households were recorded using Global Positioning System 60 (GPS) units (GARMIN, Gsmap, 64sx) with an accuracy of 5 meters. The houses selected for the entomological study and the main breeding sites of vectors were georeferenced. The data on Plasmodium density, IgG level to the gSG6-P1 salivary peptide and the Anopheles HBR were spatially linked to the GPS data of each household. Once these results were entered into the attribute table of the QGIS software, different intervals were defined by equal interval classification for mapping purposes. Parasites density was identified by the brown color. The smallest densities ranged from 56 parasite/µL blood to 1,500 parasite/µL blood and were represented by points of 1.8 thickness, the averages from 1,500 parasite/µL to 2,0000 parasite/µL with 2.4 thickness, and the highest from 20,000 to 124,160 with a square of 4 thickness. For the exposure, the fluorescent green color was chosen, and an equal interval classification from 0 to 1 was applied. These procedures considered both the dry and rainy season. Finally, the HLC site location were plotted in the map. An interpolation using the inverse distance weighting method was carried out with pixels of equal columns and rows of 254 for a better rendering of the capture points. The background colors represent the intensity of HBR from blue for the lowest rates, to orange for mild exposure, and to red for the highest rates.

Additionally, the centroid of each village was derived using QGIS 3.20 software to validate the recorded data from the portable GPS units. All relevant geographic entities were reprojected onto the WGS_1984_UTM_Zone_30N coordinate system to integrate the dataset into a geodatabase, a spatial data structure used in QGIS.

Measuring household distance to the lowland

The measurement of the distance between household and the nearest lowland was conducted using a GPS. During this phase, the GPS automatically recorded the traveled path as a line and a succession of points. A few selected households from each site and their surroundings were digitized by visual interpretation of a Bing satellite image on the QGIS 3.20 software. For the other households, the distance was measured directly to obtain a real distance. The measurement of distance from lowlands to households on the map (cm) was divided by the distance on the ground that corresponds to 1 cm. We were also able to express this scale as a fraction, where 1 cm represents 200m.

Statistical analysis:

A database was developed using Microsoft Excel software and then the data was analyzed with Graph Pad Prism 9 (Graph Pad Software, San Diego, USA). After checking that our data did not follow a Gaussian distribution (normal distribution), the non-parametric Mann-Whitney U-test was used to compare the IgG levels between two independent groups, and the non-parametric Kruskal-Wallis test was used for comparison of the IgG levels between more than two groups and the Dunn’s post-hoc test was also performed for two-by-two comparisons in more than two groups. The paired Wilcoxon test was used to compare IgG response levels between the two time points in each setting or site. The Spearman’s rank correlation test was used to analyze the correlation between the OD of tested sera with the distance of the household to the nearest lowland. The HBR was compared between sites by using the prop.test testing the null hypothesis that the proportions in several groups are the same. All differences were considered significant at p < 0.05.

Study population characteristics, parasitological and entomological data according to setting and sites

Characteristics of the study population (gender ratio and age), parasitological (prevalence and geometric mean of P. falciparum density) and entomological data (Anopheles Human Biting Rate) are presented according to study sites and season in (Table 1).

Table 1

Characterization of populations, prevalence, parasite density of Plasmodium falciparum and *Anopheles* human bite rate.
Table 1: Characterization of populations, prevalence ,parasty density of Plasmodium falciparum and Anopheles human bite rate
					Dry season			Rainy season
		n	Mean age	Gender ratio	Parasite prevalence (%)	Parasite density (95%CI)	Anopheles HBR (bpn)	Parasite prevalence (%)	Parasite density (95%CI)	Anopheles HBR (bpn)
		n	(95%CI)	Gender ratio	Parasite prevalence (%)	Parasite density (95%CI)	Anopheles HBR (bpn)	Parasite prevalence (%)	Parasite density (95%CI)	Anopheles HBR (bpn)
Rural setting	Assoum1	46	10.81	1.55 (28/18)	18 (39.13)	396.93 (125–670)	64.12	20 (43.47)	7675.30 (7403–7947)	109
	Assoum1	46	(6,54 − 15,09)	1.55 (28/18)	18 (39.13)	396.93 (125–670)	64.12	20 (43.47)	7675.30 (7403–7947)	109
	Bromakote	54	9.52	0.86 (25/29)	30 (55.55)	3232 (1504–4959)	27.25	30 (55.55)	4942.93 (3215–6670)	58
	Bromakote	54	(6.69–12.36)	0.86 (25/29)	30 (55.55)	3232 (1504–4959)	27.25	30 (55.55)	4942.93 (3215–6670)	58
	Teguiduo	52	7.18	0.73 (22/30)	30 (57.69)	3193.17 (425–5961)	66.37	32 (61.53)	2858.65	67
	Teguiduo		(4.98–9.38)	0.73 (22/30)	30 (57.69)	3193.17 (425–5961)	66.37	32 (61.53)	(90-5927)	67
subtotal		152		0.97 (75/77)	78 (51.31)	2360.73	52.58	82 (53.95)	5056.79 (2519–7594)	78.3
			9.11			(1217–3504)	(IC = 0-125)
			(7.30-10.92)			(1217–3504)	(IC = 0-125)
Urban setting	Gborotchara	44	10.64	1.20 (24/20)	7 (15.90)	472.70 (0-1278)	2.75	9 (20.45)	372 (195–549)	55
	Gborotchara		(6.85–14.44)	1.20 (24/20)	7 (15.90)	472.70 (0-1278)	2.75	9 (20.45)	372 (195–549)
	Dassikeledougou	46	10.65	0.77 (20/26)	12 (26.08)	1495.02 (76-2914)	4.12	7 (15.21)	269.56 (0-1688)	54.25
	Dassikeledougou		(6.77–14.54)	0.77 (20/26)	12 (26.08)	1495.02 (76-2914)	4.12	7 (15.21)	269.56 (0-1688)
	Doropo2	56	7.67	1.77 (28/26)	9 (16.66)	470.24 (37–903)	77,5	11 (19.64)	1363.33 (930–1796)	115.25
	Doropo2		(4.66–10.70)	1.77 (28/26)	9 (16.66)	470.24 (37–903)	77,5	11 (19.64)	1363.33 (930–1796)
subtotal		146		1.13 (82/72)	28 (19.17)	798.35	28,12	27 (18.49)	711.02 (144–1277)	58,42
		146	9.53			(256–1340)	(IC = 1-113)			(IC = 0-185)
			(7.50-11.68)			(256–1340)	(IC = 1-113)			(IC = 0-185)
TOTAL	mean	296	9.32	0.98 (147/149)	106 (35.81)	1600.65	40.35	109 (36.82)	2942	68.37
TOTAL	mean		(7.96–10.68)	0.98 (147/149)	106 (35.81)	(952–2249)	40.35	109 (36.82)	(1590–4295)	68.37

In total, 296 participants aged from 0 to 80 years (mean age = 9.32; 95% confidence interval (95%CI = 7.96–10.68) were included in the study. Mean age of the participants was similar between study sites (p = 0.22).

In rural setting, the parasite prevalence was similar between the dry and the rainy season (p = 0.646) with an average of half of the population presented an infection to P. falciparum. According to village, in dry season it ranged from 39,1% (Assoum) to 57,7% (Téguiduo) and from 43,5% (Assoum) to 61,5% (Téguiduo) in rainy season. The malaria prevalence was not significantly different between the 3 rural sites in dry (p = 0.137) and rainy season (p = 0.193) and the prevalence was similar in each site between dry and rainy season. Globally, geometric mean parasite density was significantly different between the dry and the rainy season (p = 0.019). It was different between the 3 rural sites only during the dry season (p = 0.036) and two-by-two comparison using Dunn’s test indicated a statistical difference between Assoum1 and the other sites. The geometric parasite means in Assoum1 significantly increased between the dry and the rainy season (p = 0 < 001) while it was not significantly different in Bromakote (p = 0.928) and Téguiduo (p = 0.438). A high mean level of Anopheles bites was recorded in rural setting in the dry season (52.6 bites per human per night, BHN) and in the rainy season (78.3 BHN). The aggressivity of An. gambiae was higher in Assoum1 and Téguiduo in the dry season with about 65 BHN while 27 BHN was recorded in Bromakote. During the rainy season, the HBR significantly increased (double) to 109,5 BHN in Assoum (p < 0.001), and to 58 BHN in Bromakote (p = 0.001), while it stayed similar in Téguiduo.

In urban setting, epidemiological indicators of malaria transmission were lower, with a mean of parasite prevalence less than 20% in the dry and rainy season. In dry season, the highest Plasmodium prevalence was observed in Dassikeledougou (26.1%) while the sites of Gborotchara and Doropo presented a prevalence of 16% but the difference between the 3 urban sites was not significant (p = 0.384). In the rainy season, the prevalence was also similar between the 3 urban sites (p = 0.758) with Gborotchara (20.45%) and Doropo (19.6%) presented the highest rate. Geometric parasite mean was similar between the 3 urban sites during the dry (p = 0.842) and rainy season (p = 0.705). No significant changes in the geometric parasite mean were noted between the dry and the rainy season in the urban setting and in each site. The HBR of An. gambiae was also globally lower than in rural setting with approximately 5 BHN recorded in Gborotchara and Dassikeledougou in the dry season and about 55 BHN in rainy season. Doropo presented the highest An. gambiae HBR both in the dry (77.5 BHN) and in the rainy (115.25 BHN) season. Anopheles aggressiveness significantly increased in all urban sites between the two seasons (p < 0.01).

IgG response to gSG6-P1 peptide according to setting and season

The specific IgG response level to the Anopheles gSG6-P1 salivary peptide represents an individual proxy of the level of exposure to Anopheles bites and was assessed in participants in each site during the dry and the rainy season. Participants presented a wide range in ∆OD_gSG6−P1 within and between the different sites, both in the dry and the rainy season (Fig. 2).

In rural setting, the IgG responses were significantly different between the 3 villages both in dry and rainy season (p < 0.0001) (Fig. 2A 2B). During the two seasons, there was no significant difference in IgG median level between the site of Assoum1 and Bromakoté. The IgG median level in Téguiduo was significantly lower to Assoum and Bromakoté both in the dry and the rainy season (p < 0.0001, Dunn’s test two-by-two comparison). A significant increase of IgG levels between the dry and the rainy season was recorded only in Téguiduo site (p = 0.037, Wilcoxon test).

In urban setting, the IgG responses were significantly different between the 3 neighborhoods both in the dry and the rainy season (p < 0.0001) (Fig. 2C 2D). Dassikeledougou presented a significantly lower IgG level to gSG6-P1 antigen compared to Doropo2 and Gborotchara both in the dry and the rainy season (p < 0.0001 and p < 0.005, respectively, Dunn’s test two-by-two comparison). Gborotchara showed a significant lower IgG level to gSG6-P1 antigen compared to Doropo2 in the dry season (p = 0.0017) while a similar median level of IgG response was recorded in rainy season (p = 0.111, Mann Whitney Test).

IgG response to gSG6-P1 peptide according to agricultural practices (irrigated / non-irrigated)

The cultures practiced by the population have been categorized into two groups: irrigated crops and non-irrigated crops. Irrigated crops were composed of market gardening, rice, and cereals (millet, corn), while non-irrigated crops included shea, cashew, coffee, cocoa, and yam.

Considering the 2 settings together, IgG median level responses to gSG6-P1 peptide was significantly higher in participants practicing irrigated crops compared to those producing non-irrigated crops, both during the dry (p = 0.011) and rainy season (p = 0.006) (Fig. 3A).

The same observation was noted in the rural setting (p = 0.035 in the dry season and p = 0.0008 in the rainy season) (Fig. 3B), while a similar IgG median level was noted in the urban context between participants practicing irrigated or not irrigated crops in the dry and in the rainy season (Fig. 3C). In both settings, people practicing irrigated or non-irrigated crops presented similar level of anti-gSG6-P1 IgG level between the dry and the rainy season.

IgG response to gSG6-P1 peptide according to distance between household and wetlands

The distance of each household from the nearest wetland was measured and households were then categorized in two groups according to the distance: households located less than 400m from the wetland, or more than > 400m (Fig. 4).

In rural setting, no significant difference in anti-gSG6-P1 IgG level was noted between the two groups of households during the dry (p = 0.143) and the rainy season (p = 0.096) (Fig. 4A). Only the households located more than 400m to the wetland presented a significantly different IgG level in the rainy season compared to the dry season (p = 0.044). In urban setting, households close to the wetland presented a significantly higher level of IgG response compared to the households far from the wetland both in the dry (p < 0.0001) and the rainy season (p < 0.0001) (Fig. 4B). No significant difference was noted for each group of households between the dry and the rainy season.

Correlation between IgG level to gSG6-P1 and distance from households to lowlands.

The correlation between the IgG levels specific to the salivary antigen and the distance to wetland was compared for each participant individually in the two settings separately using a Spearman's rank correlation test, and the corresponding p-values were determined (Fig. 5).

A moderate significant negative correlation between the anti-gSG6-P1 IgG level and the distance to wetland was found only in urban setting in the dry (r=-0.53, p < 0.0001) and in the rainy season (r=-0.44, p < 0,0001). No correlation was found in the rural setting between the IgG level and the distance to wetland.

Season maps of risk of exposure to An. gambiae bites

Maps were generating by combining indicators of malaria transmission (anti-gSG6-P1 IgG response, HBR and mean of parasite density in each study site during the dry and the rainy season (Supplementary data, Fig. 1). Maps indicated variation in the spatial distribution of the IgG levels to the gSG6-P1 peptide and the levels of parasitemia across space and season according to the sites. In Bromakoté, there was a high homogeneous risk of exposure to Anopheles bites and infected households were scattered throughout the village both during the dry and the rainy season, as observed in the urban setting Doropo2 and the rural site of Téguiduo where a majority of infected households being dispersed throughout the area during the two seasons.

In some villages the highest level of malaria indicators were observed in the same households during the two seasons. For example, in Assoum most households were at a high risk of exposure to Anopheles bites, and infected households were clustered in a patch in the northern part of the village during the two seasons. In the urban setting, Gborotchara is a highly urbanized neighborhood with no presence of wetland. The risk of exposure and infection was low during the two seasons, with the highest exposed households located in the western part of the site while the infected household were in a patch in the centre.

Maps also indicated variation in the spatial distribution of the IgG levels to the gSG6-P1 peptide and the levels of parasitemia across space and season in Dassikélédougou. Although the wetland was not exploited, market gardening was practiced on its edge. The highly exposed and infected households were in different place according to the seasons.

This work aims to explore the transmission of malaria in an agricultural context and to specifically investigate the impact of the presence of wetland areas and agricultural practices on the heterogeneity of malaria transmission in rural and semi-urban sites. This study aims to gain a deeper understanding of the environmental factors that contribute to the spread of malaria in the health district of Bouna, Côte d’Ivoire. In the present study, we investigated different epidemiological parameters which were Plasmodium falciparum prevalence and density, Anopheles aggressiveness and the IgG response to the gSG6-P1 salivary peptide that represents a proxy of exposure to Anopheles bites.

Exposure to Anopheles bites was thus investigated via two complementary methods giving different level of information about the exposure to Anopheles bites. First, the entomological indicator using the human landing catches (HLC) that indicates the mean number of bites that an individual may receive per night, appreciates as an approximate proxy the level of exposure to Anopheles at site scale. Second, we use a serological biomarker of exposure based on the quantification of the IgG response specific to the An. gambiae gSG6-P1 salivary peptide to evaluate the human–Anopheles contact at the individual level. This tool integrates the individual risk factors of being bitten and provides for each participant a proxy of the individual level of exposure to Anopheles bites [8]. It is more appropriate for reflecting the inter-individual heterogeneity of exposure (attractiveness to mosquito, use of personal protection as bednet for example) in natural setting [9]. Numerous studies have evidenced the Anopheles gSG6-P1 salivary peptide represents a reliable and complementary tool to entomological methods for assessing spatial and temporal heterogeneity of exposure to Anopheles bites at the individual level[16].

In the present study, malaria transmission was heterogeneous between the 6 studied sites, with a prevalence of Plasmodium infection ranging from 15% to nearly 60% in dry season and in rainy season according to site. Malaria transmission was higher in rural sites where almost half of the population was infected compared to urban neighborhoods. Exposure to Anopheles bites was also heterogeneous in Bouna during the study, ranging from 2.7 bites per night (bpn) to 77.7 bpn in the dry season and from 54.2 bpn to 115.2 bpn in the rainy season, according to site.

Plasmodium prevalence was similar in each site between the dry and the rainy season both in the rural and the urban setting, indicating the malaria transmission is perennial throughout the year in the city of Bouna and in its surroundings. It may also suggest that persistent asymptomatic infections occurred in the urban setting, where only few Anopheles were caught during the dry season. The presence of wetland and the agricultural practices (irrigated rice culture and market gardening) could create potential breeding sites for Anopheles mosquitoes during the dry season that may sustained exposure to Anopheles and thus malaria transmission throughout the year. In the rainy season, new temporary Anopheles breeding sites could be created with the water rains and came to be added to the permanent breeding sites that wetland could represent. In the rural setting, entomological studies reported the presence of a high density of Anopheles mosquitoes throughout the year in the rural villages with a mean of 52 bpn and 78 bpn recorded during the dry season and the rainy season, respectively. We also noticed that Bromakote and Gborotchara, the two sites without the presence of lowlands, had the lowest density of Anopheles for each setting, both in the dry and the rainy season.

To decipher the role of agricultural environment and practices in the risk of malaria exposure, we investigated the variation of the IgG level to the gSG6-P1 salivary peptide according to the surveys (dry and rainy season) and to the site. We first observed that the type of culture influenced the individual exposure to Anopheles bites. In rural setting, individuals that practiced irrigated culture (rice, corn, market gardening) presented a significant higher exposure to Anopheles compared to individuals practicing non–irrigated cultures. In rural setting, where irrigated crops or market gardening are commonly grown near households, high exposure to mosquito bites was observed, this result corroborate with other study realized by [17, 18]. In urban neighborhoods, there is limited space between dwellings and cultural practices are restricted in peri-urban districts [19]. This is reflected by the low number of individuals who reported practicing irrigated culture in urban setting. In addition, the exposure to Anopheles bites increased significantly from the dry to the rainy season only in individuals from rural setting who practiced non-irrigated culture. This highlighted the role of irrigated crops in the sustainability of malaria risk transmission in providing breeding sites for Anopheles mosquitoes throughout the year. Secondly, we noticed a significant negative correlation between the level of IgG level and the distance of the households of participants to the nearest wetland in urban setting. This evidenced that proximity to wetlands, which are often associated with agricultural activities, may increase the risk of malaria transmission. This is consistent with previous studies that have identified the role of agricultural activities and the presence of stagnant water in promoting malaria transmission [4, 20].

Then, we mapped the different epidemiological indicators of malaria transmission on maps for the dry and the rainy season with the objective to identify households or groups of households that presented a high risk of malaria transmission throughout the year, namely hotspot of transmission. We identified mainly two types of hotspots that differed according to their spatial dynamic. One type is represented by micro-geographical areas within a study site, that is composed of a group of households where the risk of exposure and infection was persistent between seasons. In Assoum and Gborotchara, malaria hotspots were composed of a group of infected households concentrated in a patch, persisting during the two seasons and from which transmission spread during period of high transmission with surrounding households infected in rainy season. In Téguiduo, the households the most at risk were scattered throughout the site and where almost the same during the two seasons. In Bromakoté, identifying and targeting hotspots would be crucial for effective malaria control.

Therefore, effective management of wetlands and implement Integrated Vector Control Management strategies alongside irrigation practices could help to reduce malaria transmission in endemic areas.

This study highlights the importance of considering local environmental factors, such as proximity to wetlands, in the assessment and management of malaria risk. Targeted interventions that consider the specific factors of risk in different settings would be more effective in reducing malaria transmission and ultimately improving public health outcomes [21].

About management, in neighborhoods where the risk is high or in peri-urban neighborhoods such as Doropo2, seasonal chemoprophylaxis can be initiated where two seasons are well distinguishable which is a comfortable asset for carrying out chemoprophylaxis.

For the lowlands exploited in the peri-urban districts and in the villages, a harmonization of the cultures is essential to better control the proliferation and to manage the resistance to the insecticides related to the pressure of selection emanating from the anarchic use of the pesticides especially in the fields.

In conclusion, this study investigated the transmission of malaria in an agricultural context, focusing on the influence of wetlands and agricultural practices on malaria heterogeneity in rural and semi-urban areas. The findings showed different malaria transmission rates across rural and semi-urban sites, with higher prevalence in rural settings. Exposure to Anopheles mosquito bites also varied significantly within the study area. The presence of wetlands and agricultural activities, particularly irrigated crops, played a role in sustaining malaria transmission throughout the year. Proximity to wetlands increased the risk of malaria transmission in urban areas. Mapping efforts identified both micro-geographical and larger hotspot areas with persistent transmission. This study underscores the importance of considering local environmental factors for effective malaria risk assessment and management. Targeted interventions and hotspot targeting can contribute to reducing malaria incidence and facilitate malaria elimination efforts.

Ethics approval and consent to participate

Consent for publication

Availability of data and materials

Competing interests

Funding

Authors' contributions

Acknowledgements

We acknowledge population from Bouna, community, their regional and departmental health authorities. We would like to thank the personnel of CSRS, NIHP/IPR and IRD. The authors also thank the other team members for their assistance during data collection. With the support of labEx CeMEB, an ANR "Investissements d'avenir" program (ANR-10-LABX-04-01).

Funding

This study is a part of the project "Management of vector control in agricultural ecosystems in Côte d’Ivoire”, received financial support from The Global Fund to Fight AIDS, Tuberculosis and Malaria.

Availability of data and materials

The data that support the findings of this study are available from NMCP of Cote d’Ivoire, but restrictions apply to the availability of these data, which were used under license for the current study, and so are not publicly available. Data are however available from the authors upon reasonable request and with permission of NMCP.

Authors’ contributions

OAHB, APHK, SBA, AP, GBK and MATdesigned the project. OAHB, KJDD and APHKcollected data. OAHB, SBA, AP and MD collated the data, produced all the figures analyzed the data and wrote the paper. All authors discussed results, commented on, and approved the final manuscript.

Consent for publication

Not applicable

Competing interests

The authors declare that they have no competing interests.

Ethics statement

This study followed the ethical principles recommended by the Edinburgh revision of the Declaration of Helsinki and was approved by the Ethics Committee of the Ministry of Health of Côte d’Ivoire (June 2021; No. 021/MSHP/CNSVS-kn). Written informed consent of all participants or parents of children who participated in the study was obtained before inclusion.

OMS. Principaux messages: rapport 2022 sur le paludisme dans le monde. 2022 [cited 5 Apr 2023]. Available: https://www.who.int/fr/publications/m/item/WHO-UCN-GMP-2022.07
Dossou-Yovo J, Ouattara A, Doannio JMC, Rivière F, Chauvancy G, Meunier J-Y. Aspects du paludisme dans un village de savane humide de Côte d’Ivoire. Med Trop. 1995. vol. 54, p. 331-6
Barrack O-OJ, Zhou G, Yan G, Mushinzimana E, Minakawa N, Githeko AK, et al. association between land cover and habitat productivity of malaria vectors in western kenyan highlands. Am J Trop Med Hyg. 2006;74: 69–75. doi:10.4269/ajtmh.2006.74.69
Klinkenberg E, McCall PJ, Hastings IM, Wilson MD, Amerasinghe FP, Donnelly MJ. Malaria and Irrigated Crops, Accra, Ghana. Emerg Infect Dis. 2005;11: 1290–1293. doi:10.3201/eid1108.041095
Dossou-yovo J, Doannio J, Rivière F, Duval J. Rice cultivation and malaria transmission in Bouaké city (Côte d’Ivoire). Acta Trop. 1994;57: 91–94. doi:10.1016/0001-706X(94)90097-3
Afrane YA, Klinkenberg E, Drechsel P, Owusu-Daaku K, Garms R, Kruppa T. Does irrigated urban agriculture influence the transmission of malaria in the city of Kumasi, Ghana? Acta Trop. 2004;89: 125–134. doi:10.1016/j.actatropica.2003.06.001
Bejon P, Williams TN, Liljander A, Noor AM, Wambua J, Ogada E, et al. Stable and Unstable Malaria Hotspots in Longitudinal Cohort Studies in Kenya. PLOS Med. 2010;7: e1000304. doi:10.1371/journal.pmed.1000304
Poinsignon A, Cornelie S, Mestres-Simon M, Lanfrancotti A, Rossignol M, Boulanger D, et al. Novel Peptide Marker Corresponding to Salivary Protein gSG6 Potentially Identifies Exposure to Anopheles Bites. PLOS ONE. 2008;3: e2472. doi:10.1371/journal.pone.0002472
Traoré DF, Sagna AB, Adja AM, Zoh DD, Adou KA, Lingué KN, et al. Exploring the heterogeneity of human exposure to malaria vectors in an urban setting, Bouaké, Côte d’Ivoire, using an immuno-epidemiological biomarker. Malar J. 2019;18: 68. doi:10.1186/s12936-019-2696-z
Ya-umphan P, Cerqueira D, Parker DM, Cottrell G, Poinsignon A, Remoue F, et al. Use of an Anopheles Salivary Biomarker to Assess Malaria Transmission Risk Along the Thailand-Myanmar Border. J Infect Dis. 2017;215: 396–404. doi:10.1093/infdis/jiw543
Sagna AB, Gaayeb L, Sarr JB, Senghor S, Poinsignon A, Boutouaba-Combe S, et al. Plasmodium falciparum infection during dry season: IgG responses to Anopheles gambiae salivary gSG6-P1 peptide as sensitive biomarker for malaria risk in Northern Senegal. Malar J. 2013;12: 301. doi:10.1186/1475-2875-12-301
Drame PM, Diallo A, Poinsignon A, Boussari O, Santos SD, Machault V, et al. Evaluation of the Effectiveness of Malaria Vector Control Measures in Urban Settings of Dakar by a Specific Anopheles Salivary Biomarker. PLOS ONE. 2013;8: e66354. doi:10.1371/journal.pone.0066354
PNLP. Rapport de campagne de distribution de masse MILDA. 2021.
Assouho KF, Adja AM, Guindo-Coulibaly N, Tia E, Kouadio AMN, Zoh DD, et al. Vectorial Transmission of Malaria in Major Districts of Côte d’Ivoire. J Med Entomol. 2019. doi:10.1093/jme/tjz207
Gillies MT, De Meillon B. The Anophelinae of Africa south of the Sahara (Ethiopian Zoogeographical Region). Anophelinae Afr South Sahara Ethiop Zoogeographical Reg. 1968 [cited 19 Jul 2022]. Available: https://www.cabdirect.org/cabdirect/abstract/19691000155
Wikel SK, Aksoy S, Dimopoulos G, editors. Arthropod vector: controller of disease transmission. London ; San Diego, CA: Elsevier/Academic Press, and imprint of Elsevier; 2017.
Ijumba JN, Mosha FW, Lindsay SW. Malaria transmission risk variations derived from different agricultural practices in an irrigated area of northern Tanzania. Med Vet Entomol. 2002;16: 28–38. doi:10.1046/j.0269-283x.2002.00337.x
Ijumba JN, Lindsay SW. Impact of irrigation on malaria in Africa: paddies paradox. Med Vet Entomol. 2001;15: 1–11. doi:10.1046/j.1365-2915.2001.00279.x
Jaleta KT, Hill SR, Seyoum E, Balkew M, Gebre-Michael T, Ignell R, et al. Agro-ecosystems impact malaria prevalence: large-scale irrigation drives vector population in western Ethiopia. Malar J. 2013;12: 350. doi:10.1186/1475-2875-12-350
Keiser J, Maltese MF, Erlanger TE, Bos R, Tanner M, Singer BH, et al. Effect of irrigated rice agriculture on Japanese encephalitis, including challenges and opportunities for integrated vector management. Acta Trop. 2005;95: 40–57. doi:10.1016/j.actatropica.2005.04.012
Bousema T, Stevenson J, Baidjoe A, Stresman G, Griffin JT, Kleinschmidt I, et al. The impact of hotspot-targeted interventions on malaria transmission: study protocol for a cluster-randomized controlled trial. Trials. 2013;14: 36. doi:10.1186/1745-6215-14-36

No competing interests reported.

SupplementarydataFig.1.docx
Supplementary data Fig.1: maps of malaria risk between the dry and the rainy season

Download PDF

Version 1

posted

You are reading this latest preprint version

Implication of agricultural practices in the micro-geographic heterogeneity of malaria transmission in Bouna, Côte d’Ivoire

Status:

Version 1

Abstract

Background

Methods

Results

Conclusions

Figures

BACKGROUND

METHODS

Study sites

Study Design, Procedure, and Sample Collection

Mosquito collection

Evaluation of the use of LLINs and inventory of agricultural practices

Maps

Measuring household distance to the lowland

Statistical analysis:

RESULTS

Study population characteristics, parasitological and entomological data according to setting and sites

IgG response to gSG6-P1 peptide according to setting and season

IgG response to gSG6-P1 peptide according to agricultural practices (irrigated / non-irrigated)

IgG response to gSG6-P1 peptide according to distance between household and wetlands

DISCUSSION

Conclusions

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1