Probing the composition of Plasmodium species contained in malaria infections in the Eastern region of Ghana

Malaria, a parasitic disease caused in humans by five different species of the Plasmodium genus namely P. falciparum, P. malariae, P. ovale, P. vivax and P. knowlesi, remains a devastating public health problem in the tropical and subtropical regions worldwide [1]. The burden of malaria in Ghana remains high, despite the decrease in malaria related deaths from 2985 to 1264 in 2016 [2]. Notwithstanding the slowdown in the prevalence of symptomatic malaria [1], afebrile carriage of malaria parasites is a huge problem in a number of malaria endemic countries, mainly because the parasite carriers do not exhibit any of the clinical symptoms of malaria and are thus untreated [3]. Since gametocytes, the transmissible forms of the malaria parasite are produced at each erythrocytic cycle of the parasite these asymptomatic carriers continuously serve as transmission reservoirs [4] until they are cured of their infection.

In sub-Saharan African, the majority of malaria cases are caused by P. falciparum [1] with a minor but underestimated prevalence of other Plasmodium species [5]. In Ghana, P. falciparum is the most prevalent malaria causing species with a prevalence of 98% followed by P. malariae and P. ovale with prevalence of 2–9 and 1% respectively [6]. A study by Owusu et al in 2017, reported P. malariae prevalence of 12.7% in the Kwahu south Region of Ghana [5] as compared to the national P. malariae prevalence of 2–9% [6]. The global distribution of P. malariae is sparse and variable, but is similarly endemic to West Africa, and other malaria endemic areas of the world [7‐9]. Plasmodium malariae infections usually present as asymptomatic infections although some may result in clinical disease state [10, 11]. The distribution of P. ovale is relatively limited but highly prevalent in tropical areas of Africa, including sub-Saharan Africa [12]. Plasmodium vivax is endemic in Asia but scarce in West Africa where the natives lack the Duffy antigen receptor for chemokines, an essential receptor for erythrocyte invasion by P. vivax [13]. Despite the widespread absence of the Duffy antigen receptor for chemokines in people from sub Saharan Africa a few recent reports of have identified P. vivax infections in some sub Saharan African countries including Mali and Nigeria [14‐16].

In order to implement accurate measures for effective control and treatment of malaria the detection of all human Plasmodium species is important [17]. Recently, the definition of malaria elimination has been revised to include the interruption of the local transmission of all human malaria parasites [18], making it necessary that national control programs include surveillance of all malaria parasite species. However, due to the very low occurrence and parasite densities of P. malariae and P. ovale in sub-Saharan Africa where P. falciparum is endemic, only a few microscopists are able to correctly identify P. malariae and P. ovale infections [19]. This combined with the fact that most P. malariae and P. ovale infections present as mixed infections with P. falciparum [11, 20], likely contribute to the misdiagnosis and the very low reported prevalence of other Plasmodium species in populations where P. falciparum is highly endemic. Rapid diagnostic test kits for malaria have improved malaria diagnostics, however this is mainly true for P. falciparum infections where the histidine rich protein 2 (HRP2) antigen, which is specific for P. falciparum, is detected as these kits have the highest sensitivities compared with the other rapid diagnostic kits for malaria [21]. Detection of parasite lactate dehydrogenase and aldolase antigens can be used for the detection of all Plasmodium species either combined as a pan specific test kit (aldolase) or separately using species-specific lactate dehydrogenase, and these have higher specificities than the HRP2 based tests but are less sensitive [21].

In this study we sought to determine whether the contribution of P. malariae and P. ovale to the overall prevalence of malaria captured during multiple community surveys in the Eastern Region of Ghana could be accurately predicted without the use of molecular tools, especially as numerous community surveys are carried out without the use of molecular tools. As such we used microscopy, RDT and species-specific PCR to assess the prevalence and composition of malaria parasites carried by consenting individuals living in seven closely linked communities in the Eastern Region of Ghana.

The study population comprised of 729 participants aged below 85 years (Table 1) recruited from 7 closely linked communities. The proportion of males in the study population ranged between 37% (17/46) in young adults between 15 years to 19 years and 60% (74/123) in children between 5 years to 9 years (Table 1). A total of 3.4% (25/729) of the study participants were identified as febrile and had axillary temperatures of or above 37.5 °C. Children aged between 5 and 9 years had the highest prevalence of febrile cases (5.6%) and adults ≥20 years old had the least prevalence of febrile cases (2.4%) (Table 1). There was no significant difference (Kruskal-Wallis test, p = 0.6621) in the median ages of the participants from all the 7 communities.

In most malaria endemic countries in the World Health Organization Africa Region, P. falciparum is the major malaria parasite and represents close to 90% or more of the total malaria parasite population [1]. The 2018 world malaria report estimated P. falciparum to represent 100% of the total parasite population in Ghana [1]. Similarly, the National Malaria Control Program, Ghana uses parasitaemia in referring to the presence of malaria parasites, without the distinction between the various Plasmodium species [2, 28‐30]. Co-infections of P. malariae or P. ovale with P. falciparum have been reported to lower the total parasite density of an infection [10] and also increase P. falciparum gametocyte prevalence [31]. Very few reports have determined the distribution or prevalence of non-falciparum parasites [5, 32], making this study very essential not only to fill a knowledge gap but also to inform policy on the need for sensitive tools to enable the identification of the under reported prevalence of P. malariae and P. ovale parasites.

The prevalence of Plasmodium carriers was generally high in this study with a PCR estimated average of 66% and a microscopy average of 32%. The increased prevalence determined by PCR is expected as microscopy is known to underestimate parasite prevalence, especially at low parasite densities [33]. The high prevalence of submicroscopic infections identified in this study as well as other studies [33‐35] is a big challenge to malaria control in Ghana as the Test Treat and Track program is based on the microscopic detection of malaria parasites. This suggests that even if the program is 100% effective, malaria transmission will still be very high as half of the infected population is afebrile, thus not identified and treated. In this study, the prevalence P. malariae estimated by microscopy and PCR were similar, which was contrary to the differences between P. falciparum and P. ovale estimates, which were much higher by PCR when compared with microscopy. One main reason for the disparity was the high number of P. falciparum samples that were misclassified as P. malariae by the microscopists. Misclassification of Plasmodium species is particularly common in countries with a low prevalence of non-falciparum malaria primarily because of infrequent encounter by the microscopist and the subsequent low ability to identify morphological differences amongst the various Plasmodium species [36]. The World Health Organization recommends that uncomplicated infections harboring P. falciparum and P. malariae are treated with the same dose of artemisinin combination therapy [17] as such, a misclassification of P. malariae for P. falciparum and vice versa may not result in severe outcomes. However, due to the requirement of a radical cure agent such as Primaquine to clear P. ovale hypnozoites the treatment regimen is artemisinin combination therapy + Primaquine [17], thus misclassifications of P. ovale with P. falciparum could lead to severe P. ovale infections.

Children under the age of 5 years are considered to be the most vulnerable groups [1], however, in this study, children under 5 years did not have the highest prevalence of infection, although they had the highest parasite load. This observation supports the results from an earlier study conducted on adults and children with asymptomatic malaria infections living in Northern Ghana [35] as well as another study conducted on children from Ivory Coast and Mauritania [37]. A contrary finding has been reported in a study conducted on symptomatic malaria patients, where the parasite load in children under 5 was lower than those of older children [38]. The disparity in these findings could be due to the use of volunteers with different malaria status and suggests that young children with afebrile malaria infections may be able to withstand higher parasite densities compared with their symptomatic counterparts and older age groups. Older children aged between 10 and 14 years old were identified as having the highest prevalence of afebrile Plasmodium carriers, including both P. falciparum asexual and gametocyte as well as P. malariae parasite carriage. High gametocyte carriage in the group with the highest prevalence of both P. falciparum and P. malariae could be due to intra-host competition, which is known to enhance P. falciparum gametocyte production [31]. An earlier report on asymptomatic carriers in Kenya also identified children below the age of 15 years to harbor the highest levels of P. malariae than the older population, which is similar to the findings in this study [39]. All three groups of children in this study are more likely to enhance malaria transmission due to the high prevalence of microscopic densities of gametocytes carried by the group members. Although membrane feedings assays were not conducted in this study, other studies have found that high-density gametocyte infections contribute more to malaria transmission than low-density (sub microscopic) gametocyte infections [40].

In this study, age influenced the false positive RDT rates, with children having much higher false positive rates than the adult group. One main reason why the three groups of children exhibited high false positive rates could be because children generally have more incidents of clinical malaria as well as have higher density infections [41], which when treated would result in high levels of HRP2 antigen and a resultant longer duration of HRP2 antigen persistence that causes false positive RDT result [42]. Persistence of HRP2 antigen that is responsible for majority of false positive HRP2 RDT results is concentration dependent [43] and will take longer to clear after the treatment of high-density infections [44, 45]. False negative RDT results have also been associated with the presence of parasites with deletions in the Pfhrp2 gene [46, 47] or the presence of very high parasite densities that result in high concentrations of HRP2 antigen and a subsequent prozone effect [48]. False negative microscopy results are mainly due to the presence of submicroscopic densities of parasites, which are very common in asymptomatic infections [33]. Though there might be a possibility of false positive PCR results due to either contaminations during sample processing and amplification [49], nucleic acid amplification methods such as PCR has been recommended by the World Health Organization to be used to provide more accurate prevalence measurements during surveillance exercises as well as for malaria parasite detection in low transmission settings [50].

Even though afebrile parasite carriage was significantly affected by age, Plasmodium parasite carriage was similar in the seven closely linked communities. This suggests that although similar malaria control interventions would be effective in closely linked communities, interventions should be age group specific. There are a number of malaria control interventions that are targeted to protect children below the age of 5 years, one of the most vulnerable groups with respect to malaria from exposure to the malaria parasite [1]. The results from this study suggest that those interventions are effective as only a small fraction of the children below 5 years old living in the communities surveyed carried malaria parasites.

Previous reports on the distribution and prevalence of non-falciparum malaria parasites in Ghana vary from less than 1% to more than 12% across different regions of the country [5, 51, 52], with no reports of the presence of P. vivax in Ghana to date [5]. In order to obtain a more accurate measure of the nationwide distribution and prevalence of non-falciparum malaria in Ghana, it is necessary that nationwide molecular surveys of all human malaria parasites be conducted on a regular basis. This will help determine the geographic distribution and prevalence of non-falciparum malaria parasites, especially P. ovale and decide whether there is a need to implement a revision to the national malaria treatment policy to incorporate a treatment regimen for P. ovale, which is presently absent in Ghana.

The high rate of misidentification of non-falciparum parasites and the total absence of detection of P. ovale by microscopy suggests that more sensitive malaria diagnostic tools including molecular assays are required to accurately determine the prevalence of carriers of non-falciparum parasites and low-density P. falciparum infections, especially during national surveillance exercises. Additionally, malaria control interventions targeting the non-falciparum species P. malariae and P. ovale parasites are needed.