This study presents the first investigation of malaria prevalence in the urban and semi-urban population of the CAR. The results indicate a high prevalence of asymptomatic malaria in CAR, with one in three individuals carrying the disease and one in ten of those having a high parasitemia. Despite local efforts to reduce malaria transmission, the disease remains the most serious threat to public health in the Central African Republic (CAR). This is highlighted by the prevalence of malaria using microscopic techniques, which shows the existence of a human reservoir and the active circulation of the parasite in the population, particularly in Bangui. The prevalence of malaria parasitaemia in Bangui, as determined by light microscopy and RDT in our study, are slightly lower than those estimated at the national level. The PCR positive proportion in our study is similar to that of previous findings in the country, estimated at 334.7 per 1000 inhabitants [4]. In our previous studies, as in many others, malaria transmission intensities vary according to epidemiological facies, with prevalence often higher in rural than peri-urban and urban areas [19]. Given that our study area, Bangui, is an urban area and also that 3/4 of the CAR population live in rural areas, the burden of malaria in rural communities would be therefore higher. The WHO has indicated that the actual number of malaria cases in the CAR is grossly underestimated, and may even be several times higher than reports suggest [20].
The malaria parasite prevalence value found in our study is higher than that of a community-based study carried out in several rural districts in other countries, including Tanzania, with 9%, 9% and 29% of malaria prevalence for microscopy, mRDT and PCR, respectively [21]. Several factors may justify the higher value of PCR compared to microscopy and mRDT, such as low parasitaemia undetectable by microscopy and mRDT, because of the higher sensitivity of PCR (2 - 5 parasites/mL) [22], the persistence of Plasmodium DNA which require careful interpretation, and correlation with microscopy and clinical presentation [23]. The value of mRDT can also be influenced by the persistence of the PfHRP2 antigen from previous infections [24].
A similar study carried out in the west of Cameroon, a country bordering the CAR, showed a prevalence of 21% and 26% by mRDT and microscopy, respectively [25]. Another study carried out in Butajira, in the south-central region of Ethiopia, revealed a malaria prevalence of 21% using the microscopy method, which is similar to our findings [26]. In addition, a community-based study in rural areas in Nigeria showed higher prevalence rates of 56.8% with mRDT and 38.6% with microscopy [27].
Our study also revealed a significant proportion of submicroscopic infection, and measuring body temperature would have made it possible to determine the proportion of asymptomatic malaria in the study population. The submicroscopic and asymptomatic malaria carriers do not seek treatment for their infection, and therefore constitute a reservoir of parasites available for transmission by Anopheles mosquitoes [28]. Moreover, long term asymptomatic carriage may represent a form of tolerance to the parasite [29]. The prevalence of submicroscopic infection (9.56%) in our study is comparable to a study carried out in 3 villages in Senegal showing values ranging from 2.89% to 12.5% [30]. However, it is lower than that of a study conducted in Tanzania (18%) [21]. The trend in parasite density in our study (geometric mean and range) are consistent with data from a community-based study in Cameroon, although carried out in children under 15 years of age [31].
Plasmodium falciparum infection was found in 98.55% and P. ovale in 3.38% only and in coinfection, consistent with existing data showing that P. falciparum is responsible for 82–100% of all malaria cases [32, 33]. This result confirms the predominant circulation of P. falciparum in CAR and the low circulation of P. ovale, which is becoming endemic, since work documented in recent years has shown that this species does circulate, although at a rate of around 2% [34, 35]. This calls for continuous monitoring of P. ovale in the CAR. The potential role of hypnozoite-induced relapses in P. ovale infection is known and widely documented [36]. However, cases of increased malaria relapse rates have been reported by clinicians but wrongly attributed to P. falciparum, leading to the abusive use of ACTs, a major cause of resistance development in circulating parasites. The national malaria management protocol needs to be revised to take account of non-falciparum malaria, the prevalence of which is increasing in Africa in general [37].
This study was conducted during the COVID-19 epidemic, which may have influenced the results. Empirical treatment with repurposed drugs such as chloroquine (CQ), hydroxychloroquine, ACT, and azithromycin, all anti-malarial drugs, was ongoing. Additionally, self-medication and traditional plant remedies are being used, which could impact malaria prevalence at the time of our study.
Furthermore, limited access to malaria diagnosis, treatment and prevention during the global COVID-19 pandemic may have contributed to the increase in the prevalence of malaria in the community [38, 39].
The prevalence of malaria was found to be higher in age groups between 15 and 49 (30%) than in other age groups in the current study, which is consistent with several studies of malaria at community level, where adults appear to be more affected than the paediatric population [26]. This could be explained by the fact that paediatric malaria is often symptomatic, and is therefore rapidly referred to health facilities. In addition, targeted malaria control interventions focusing on children under five and pregnant women, with free diagnosis and treatment, could also explain the low number of children enrolled during our study and the low prevalence of malaria.
Overall, our results highlight the microspatial heterogeneity of malaria in Bangui. Malaria prevalence was found to be greater in semi-urban boroughs (6th, 7th, and 8th) than in urban areas (1st , 2nd, 4th and 5th ) which is consistent with a recent study on malaria conducted in the network of sentinel sites set up for influenza surveillance in CAR, including Bangui, where it showed a relatively high prevalence in semiurban areas compared with urban areas [40]. Several other studies in Africa have indicated that urban areas have lower levels of malaria prevalence [19, 41]. Moreover the 4th and 5th boroughs, although urban areas have relatively the highest prevalence compared to the 1st and 2nd boroughs . These two districts are the most populated districts of Bangui and a study showed that population density in urban areas is a factor that can contribute to the persistence of malaria transmission [42]. A high urban population density is thought to have an impact on the environmental and socio-economic dimensions at the root of the vulnerability and heterogeneity of malaria in urban areas [43]. These are consistent with much of the literature on population density and urbanisation, which has shown that the risk of malaria infection still exists in densely populated urban areas of Africa, and that population density was found to be an important predictor of malaria risk [44].
Community-level malaria is an important indicator of the outcome of control programs. The identification of individuals with asymptomatic malaria via active surveillance using mRDT or PCR is an important component of malaria elimination efforts [45, 46]. The information obtained from this study can be used to improve the control and community-centered strategies. Although the study was carried out during August and September including the rainy season (May–October), this would not have had a significant impact on the result, as it has been shown that malaria transmission is perennial and therefore holo-endemic in CAR [40].
This study has limitations. To improve on the findings, it would have been beneficial to include body temperature values for the participants, allowing for a more precise definition of the prevalence of asymptomatic parasitaemia in the area. Additionally, information on the community's perception of malaria and the use of preventive measures such as MILDA and ITNs could have been useful to determine the associated factors of malaria parasite carriage in the area. It is worth noting that the mRDT Ag P.f/Pan could have provided more information on the current episode of malaria than the one we used (P.f) that also detects past infection [47].