Abstract

Background

Malaria continues to be a major health challenge globally with an estimated 655,000 to 1.2 million deaths in 2010 [1,2]. About 90% of the malaria-related deaths have been reported from sub-Saharan Africa [2]. In Mauritania, malaria has increased in recent years, but it is confined to the southern and south-eastern parts of the country, where about 60% of the resident population are at risk of infection with Plasmodium falciparum[2,3]. The nature of malaria transmission in the Senegal River Gorgol valley, southern Mauritania, is poorly understood due to the lack of epidemiological data [2-4], and also because many febrile presentations are attributed to malaria without differential diagnosis, and hence are treated presumptively [3,4]. Since “malaria cases” are reported in health facilities in the southern part of the country throughout the year, some consider the area a perennial transmission zone [3,5]. Conversely, as the area is characterized by two distinct seasons (rainy season and dry season), it is suspected that during the dry season when there is no rainfall and temperatures are very high (>45°C), malaria transmission is likely to be absent [6].

Recent entomological investigations have identified Anopheles arabiensis as the predominant vector species in the region, while An. pharoensis and An. funestus were reported in close proximity to irrigation schemes [7]. However, due to a lack of reliable health information and resources to confirm presumptive clinical cases [3,4], further investigations are needed to clarify the transmission pattern of malaria in the southern parts of Mauritania. Preceding studies revealed little or no transmission among vectors in the dry season [8] and reports on P. falciparum parasite rates (PfPR) are scarce and conflicting [3,9]. Whether laboratory-confirmed cases in the dry season are autochthonous or imported from other regions is also currently unknown, because data are collected from health facilities serving populations from various regions in Mauritania as well as neighbouring Senegal.

By studying Plasmodium parasitaemia rates in the human population in the dry season, comparisons can be made with entomological studies and laboratory data that could contribute to a better understanding of malaria transmission in the Gorgol valley. Of note, most published studies in the region on malaria are dated, they were primarily based on hospital data, and only few attempts have been made to characterize malaria infection in the preschool-aged population in the dry season. To contribute to filling in these gaps, we conducted a multipurpose study, including a determination of malaria infection among a random sample of children aged 6–59months who were recruited during a cross-sectional household survey in the dry season of 2011.

Methods

Results and discussion

As shown in Figure ​Figure1,1, 371 children were recruited in the required age range of 6–59months from 246 randomly selected households in various communes in the city of Kaédi. Figure ​Figure22 shows the geographical location of the households from which participants were recruited. The malaria prevalence rate was 0%. Indeed, neither the RDT nor the microscopic diagnosis of Giemsa-stained blood films revealed a P. falciparum infection among the 371 preschool-aged children.

Open in a separate window

Figure 1

Flow chart depicting participant recruitment and malaria test administration.

Open in a separate window

Figure 2

Map of study site with selected geo-referenced households.

It is important to note that clinical signs and symptoms of malaria are non-specific and fever is a common presentation that overlaps with other clinical entities such as respiratory tract infections and gastrointestinal illnesses [13]. We additionally enquired about the occurrence of acute respiratory infection and diarrhoea in our study sample. Hence, mothers/guardians were interviewed about these diseases using a recall period of 2weeks. We found high frequencies of acute respiratory infection and diarrhoea, 43% and 35% of the mothers/guardians interviewed reported these conditions, respectively. About two thirds of the children with acute respiratory infection and diarrhoea required further assessment by medical staff with no significant sex difference in the occurrence of these signs among participants (Table1).

Table 1

Prevalence of acute respiratory infections and diarrhoea, as determined by a questionnaire administered to mothers/guardians of children aged 6–59months in Kaédi, southern Mauritania in March 2011, stratified by sex

Disease	No. of children interviewed	No. of children tested positive (%)	No. of children needing further assessment (%)	Number (%) of children tested positive, stratified by sex		χ2	P-value
				Males (n=199)	Females (n=172)
Acute respiratory infections	371	161 (43.4)	105 (65.2)	84 (42.2)	77 (44.8)	0.25	0.620
Diarrhoea	371	130 (35.0)	87 (66.9)	63 (31.7)	67 (39.0)	2.16	0.142

Open in a separate window

Our observation of a nil PfPR concurs with recent findings obtained in entomological investigations in neighbouring regions in the dry season of 2009, as reported by Ndiath and colleagues [8]. Indeed, in the longitudinal arm of their study, the authors examined more than 2,400 Anopheles specimens during a 16-month period with an overall infection rate of 1.1%. No infected mosquitoes were observed between January and June 2009. However, some transmission was observed in January of 2010, albeit at low levels. Malaria transmission may occur between November and January because, in the absence of rainfall, the annual flood recession from the Gorgol River and the Senegal River during these months can provide residual breeding sites for mosquitoes. Our observations are also consistent with older studies in the Gorgol region conducted shortly after the filling of the Foum Glieta dam, as Baudon and colleagues reported the absence of malaria transmission in March during the dry season [9]. Rainfall time-series data for Kaédi show a high inter-annual variation, with unpredictable changes in the duration of the rainy season and amount of rainfall experienced. The absence of parasitaemia in the general population can be explained in part by the absence of rainfall coupled with high average temperatures in the city that can exceed 45°C. Clearly such climatic factors do not favour malaria transmission. In their study, Ndiath and colleagues [8] also noted a considerable difference in the number of mosquitoes caught in January 2009 (dry season) compared to October (rainy season) in the same year (27 versus 568 mosquitoes). In 2009, Dia and co-workers [7] found a low Plasmodium infection rate when they examined 394 An. arabiensis mosquitoes with only one specimen being infected with P. falciparum, owing to an infection rate of 0.25%. Furthermore, there was a considerable difference in infection rates of mosquitoes in these same months (0% compared to 1.06%), suggesting that transmission appears to occur only during the rainy season [8]. Unpredictable trends in the duration of the rainy season, and the amount of rainfall due to climate variability and change highlight the need for more appropriate public health efforts in malaria control.

Laboratory data from the Kaédi regional hospital report 4% slide positivity rate in April 2009 (unpublished observations), while Cortes and colleagues recorded a dry season prevalence of 16.6% in April-May 2003 [3]. Several factors could account for the variation between laboratory-reported estimates and findings in the population. Firstly, imported cases from neighbouring regions could account for this difference because the hospital is the only tertiary referral centre in the region, and hence it serves not only local residents, but also people from other regions in Mauritania as well as residents from neighbouring Senegal. The presence of highly focal transmission among residents who live in close proximity to irrigated rice fields or other social-ecological settings conducive for vector proliferation and transmission might also contribute to the observed variation. Even though studies have shown an association between higher anopheline densities in villages situated in flooded areas with ditch water as compared to those in sandy ground, a change in mosquito density does not automatically result in direct changes in malaria transmission [8,14]. Further in-depth entomological studies characterizing the geospatial and seasonal distribution as well as the infection rates in common vectors will undoubtedly help in better understanding the malaria transmission ‘hot spots’ in these areas.

Conclusions

The absence of Plasmodium parasitaemia among preschool-aged children in our multi-purpose cross-sectional survey conducted in Kaédi, southern Mauritania points, once more, to very little or no malaria transmission in the dry season. However, the possibility of highly focal transmission zones cannot be discounted because local transmission may occur during the annual flood recession in the months immediately following the end of the rainy season (November to January). Between February and June when the swamps have dried up and ambient temperatures are very high, acute respiratory infections and diarrhoea appear to be the underlying aetiology in most febrile illnesses among preschool-aged children. The implications are that public health efforts to control malaria should be strengthened in the rainy season, and the months immediately following the cessation of the rains. Nevertheless the predicted climate change patterns in this region require careful surveillance of malaria transmission. The case management of febrile children under the age of 5years should adopt a broader consideration of differential diagnosis, which should include acute respiratory infection and diarrhoea (among other bacterial and viral infections), as more likely aetiologies in the dry season. Importantly, malaria treatment with ACT should be strictly contingent on a positive malaria test (RDT or microscopy), which has been shown to be a safe and effective strategy to reduce unnecessary malarial treatment of febrile cases [15,16]. More research is needed to determine whether a recent travel history might explain positive laboratory-confirmed cases in health facilities in the dry season. This research will yield interesting insights about whether such cases are autochthonous or imported from other regions or from neighbouring countries, and might serve as a benchmark for investigating a potential impact of climate change.

Competing interests

The authors declare that they have no competing interests.

Authors’ contributions

ST, HB, MSW, JU and GC conceived the study. ST, HB, OB, MK and CBOAS conducted field and laboratory work. DT and IS co-coordinated the geographic survey, while MSW processed the geo-referenced data and produced the map. ST performed statistical analysis, interpreted the data and drafted the manuscript. HB, MSW, JU and GC revised the manuscript. All authors read and approved the final submitted version of the manuscript.

References