Background
Malaria remains a major endemic disease in tropical areas and a significant health threat to travellers and military personnel [
1]. Since 2003, French forces have implemented many studies to assess the malaria risk according to the locations of troops and the insecticide susceptibility of vectors in the field [
2‐
4]. In Gabon, malaria is predominantly caused by
P. falciparum and malaria transmission is perennial and was considered until today as hyperendemic [
5]. Throughout the country, infections with
P. falciparum accounted for 30% to 50% of the consultations of children with fever between 1980s and 1990s [
6,
7]. In Libreville during the same period, the prevalence of
P.falciparum infection and of anemia were respectively of 57% and 71% in pregnant women [
8] In 2005, artemisinin-based combination therapy (ACT), insecticide- treated nets (ITNs) and intermittent preventive treatment (IPT) were freely available for children under five years old and pregnant women [
9]. Since then, a decline of malaria burden has been described in Libreville both in pregnant women and children [
9,
10]. Actually, parasitaemic data in febrile children over the age of five years are questioning about both the impact on the level of transmission in Libreville and on the acquisition of malaria immunity in urban areas by children [
11]. Libreville is probably an urban area in which a shanty-town with high malaria transmission and a residential area with low transmission coexist. Unfortunately, none recent entomologic data exist on malaria transmission in Libreville or other major towns of the country. Studies conducted in the inlands showed that malaria transmission was sustained mainly by members of the
Anopheles gambiae species complex and the
Anopheles funestus Giles species complex[
5,
12,
13]. In Libreville, a study conducted during the rainy season in 2000 showed the presence of
An. gambiae s.s. molecular form S and enhanced the presence of both
kdr-e and
kdr-w alleles at high frequencies in this population [
14].
Malaria is a major health priority for French forces in Gabon since many years and always today. Many cases have occurred and many outbreaks have been described in the country or in returnees [
15,
16]. So most of the last clinical trials on alternative chemoprophylaxis for troops have been conducted in Gabon [
17,
18]. The vector control programs in the camps include personal protection measures (PPMs) and unit protection measures (UPMs). PPMs consist of proper use of uniforms, the application of repellent to exposed skin, the application of permethrin to battledress uniforms, and the use of insecticide-impregnated bed-nets. UPMs include environmental management, larviciding, and space spraying. Here are reported the results of a preliminary entomologic study conducted in Equatorial Africa in the two French military camp in Gabon: "camp de Gaulle" in Libreville, the capital city, and "camp N'tchorere" in Port-Gentil the economic capital city. The aim of this study was to assess the local malaria transmission risk for troops in these urban areas.
Discussion
Mosquito population densities were remarkable in the two camps both indoor and outdoor. This is particularly true in Port Gentil regarding the number of
Culex and
Mansonia species caught outdoors and the
An.gambiae s.l. HBR of 58.7 bites per person per night. In Libreville, most of the mosquito biting rate was due to
C.quinquefasciatus, a pest that breeds easily in waste water. Catches were particularly high near sanitary toilets suggesting that breeding sites were close in the septic tanks. This nuisance could probably be controlled with little effort in improving the airtightness of septic tanks.
Aedes population densities were low in our study. Although night captures are not appropriate for such diurnal species, these results are concordant with a specific study conducted with BG-sentinel traps to monitor the
Aedes albopictus and
Ae. aegypti population in the Libreville camp [
25].
Anopheles gambiae s.l. HBRs were low ranging from 2.3 to 16.8 bites per person per night during the study period. Figure
1 show the temporal variations in adult density and rainfall during the study. The curve of
An. gambiae s.l. HBR follow the variations of the moving average graph of the rainfall regime indicating that
An. gambiae density in Libreville is linked to the intensity of rainfall and probably to the number of temporary breeding sites. The assessment of the endophilic or exophilic behavior could have been biased by the scarcity of indoors collections and the difference between the camps have to be confirmed.
In Libreville, only one species of the
An. gambiae complex was present:
An. gambiae s.s. molecular form
S as already described [
14]. The exclusive presence of S form is consistent with the importance of rain-dependent temporary breeding sites in Libreville [
26]. In Port Gentil,
An. gambiae complex was represented by
An. melas that accounted for two thirds of catches and
An. gambiae s.s. molecular form S for the last third.
In Libreville,
An. gambiae s.s. molecular form S is the main malaria vector. According to the
An. gambiae s.s. density-rainfall association malaria transmission is expected to be lower during the dry season from June to August. Considering the results of this survey as representative of the transmission during rainy season, an estimation of a minimal mean annual EIR of 3.7 infected bite per person per year is realistic. The present observations are consistent with the results of a meta-analysis of studies of malaria transmission in sub-Saharan Africa, which found a mean annual EIR of 7.1 in the city centres, with more than two thirds of the studies reporting EIRs < 4/year [
27]. In Port-Gentil,
An. gambiae s.s. molecular form S and
An. melas were both involved in
P. falciparum transmission. If
An. gambiae s.s. molecular form S accounted for only one third of
An. gambiae s.l. bites, it was responsible of two third of the
P. falciparum transmission. Despite its lower vector competence for
P. falciparum, An. melas contribution in malaria transmission in the Port-Gentil area was high as already described [
28,
29]. Extrapolating the present findings to the nine months of the rainy season and considering a low transmission level during the dry season, the average annual EIR would be at least of 100 infected bites per person per year. The level of transmission in peri-urban Port-Gentil is similar at those encountered in rural settlements [
27]. The malaria transmission is totally different between the two French camps in Gabon. During a four months duty, each platoon of soldiers was based during two weeks in the N'tchorere camp and was clearly exposed to a high number of
P. falciparum infected bites. According to the delay of incubation, a soldier that doesn't take his chemoprophylaxis would have his malaria attack in Libreville. This situation could have lead to an overestimation of malaria burden for troops in Libreville. In other hand, few cases of malaria occurred in soldiers that were permanently stationed in Port-Gentil for two years. Perhaps the high level of nuisance, nearly 800 mosquito bites per night in February 2007, enhanced the use of personal protection measures by soldiers (i.e. repellents use, bed nets use and proper wearing of battle dress) and their compliance to malaria chemoprophylaxis. 80% of bites are occurring in the second part of the night. During this period, most of soldiers are sleeping in mosquito-proofs buildings and are so protected. Only the outside sentinels are probably exposed to mosquito bites and to malaria vectors.
As reported previously in Gabon in Libreville and in the inlands, both kdr-w and kdr-e mutations in
An. gambiae molecular form S were present [14; 30]. Contrarily to a previous study in Libreville,
kdr allele was present both in Libreville and Port-Gentil, and as already described in neighbouring countries (Equatorial Guinea and Cameroon), the kdr-w allele was the most frequent both in Libreville and Port-Gentil [
31,
32]. The distribution of the
kdr alleles was similar in the two military camps of Port-Gentil and Libreville but there was a statistically significant difference with the historical data in Libreville (Chi
2 with 2 df, P < 10
7). This difference is possibly due to sampling bias or to variations in the repartition of
kdr alleles in
An. gambiae s.s. populations between quarters in Libreville. Both
kdr-w and
kdr-e mutations have been tightly linked with DDT and pyrethroid resistance phenotypes in field populations of
An. gambiae s.s[
33‐
36]. The presence of both
kdr-e and
kdr-w alleles at high frequencies in these populations has implications for the effectiveness of the current vector control programmes, that are based on pyrethroid-impregnated bed nets both for French forces and for the Libreville or Port-Gentil inhabitants. Currently, the impact of
kdr on the efficacy of ITNs is unclear with contradictories studies [
37‐
39]. Nevertheless kdr screening is still the best molecular diagnostic tool for predicting pyrethroid and DDT efficacy in preliminary studies like this one [
40]. Therefore, before interventional studies, a standard WHO exposure to DDT or permethrin has to be conducted to clearly assess the resistance level and take into account the possible involvement of additional resistance mechanisms.
It was the first time that a molecular screening was conducted in Gabon to assess the presence of an insensitive acetylcholinesterase (AchE) in populations of
An. gambiae. A low level of insensitive AChE was detected in Libreville with an allele frequency of 0.4. None insensitive AChE was detected in Port-Gentil but there was no statistically difference in the ace1 mutation frequency between Libreville and Port-Gentil. Complementary studies have to be conducted to monitor the possible existence of a resistance to organophosphates and carbamate insecticides in urban
An.gambiae s.s. form S populations in Gabon. If carbamates remain a viable alternative in front of the low level of insensitive AchE and in the Gabonese context of a probable high resistance to pyrethroids, their extensive use alone would probably conduct to select multi-resistant specimens of
An.gambiae s.s. form S in Libreville and Port-Gentil. This possibility has to be taking in account before implementing vector control programmes both in military camps and in towns. The use of a combination of pyrethroids and organophosphates in mixture, mosaic or in parallel on different supports has been proposed against pyrethroid-resistant malaria vectors [
41‐
43]. A new mosaic long-lasting insecticidal net developed to control wild pyrethroid resistant
An. gambiae has shown no benefits compared to classical long-lasting insecticidal nets [
44,
45].
According to literature, this was the first time that entomological studies on malaria transmission were conducted in the main towns of Gabon, which represent more than 50% of the total population. Malaria transmissions exist in both towns but with high difference in the level of risk. In urban areas, malaria transmission is highly focused and a single study in one area of a town is not sufficient to assess the global level of transmission. And one should not minimize the possible impact of the vector control programme implemented in the camps all year round. In Libreville, the risk is comparable to other big African cities, but seems very low according to the equatorial situation and the previously described hyperendemic transmission. Nevertheless, these results are concordant with the clinical statements made by physicians and researchers in Libreville that all concluded to a decrease in malaria transmission since 2000 in Libreville [
9‐
11]. In Port-Gentil, the level of transmission is high but is probably not representative of the current situation in all parts of Port-Gentil.
Anopheles melas is clearly involved in malaria transmission in Port-Gentil, but further studies in different parts of Libreville are necessary to understand its involvement or not. Until today, molecular screenings of resistance in Gabon have only concerned kdr mutation. Here is the first report of insensitive acetylcholinesterase in Gabon. This paper highlights the worrying level of pyrethroids resistance of
An. gambiae s.s. for most of 50% of Gabon's population. There is a lack of entomological data to understand the dynamic of urban malaria in Gabon. Further studies have to been conducted in different areas of theses towns including central and peripheral districts, coastal and inland districts, modern and traditional districts. The situation of insecticide resistance has to be clarified in using molecular tools and insecticide bioassays in different districts of each town.
Authors' contributions
JRM was responsible for the supervision of data collection, analysis, interpretation and production of the final manuscript and revisions. TC contributed to the supervision of data collection, the data analysis, interpretation and production of final manuscript. FJ contributed to the supervision of data collection, to the data analysis. BP contributed to the supervision of data collection, to the data analysis, interpretation and production of final manuscript. RA FJ contributed to the supervision of data collection, to the data analysis. CR contributed to overall scientific management, analysis, interpretation and preparation of the final manuscript and revisions. MK contributed to overall scientific management, analysis, interpretation and preparation of the final manuscript and revisions. FP was responsible for overall scientific management, analysis, interpretation and preparation of the final manuscript and revisions. All authors have read and approved the final manuscript.