Malaria is a major debilitating infectious disease that affects millions of people and impedes public health and socioeconomic development in Africa (Kassegne et al., 2017). In tropical Africa, nearly 90% of deaths among young children are related to malaria (World Health Organization [WHO], 2018b).

Malaria transmission occurs throughout the year in Togo and is usually characterized by seasonal outbreaks with a peak during the rainy season. An estimated nearly 100% of the population of Togo lives in areas endemic for malaria—all the five administrative regions of the country having populations at risk of malaria—with disease prevalence rate estimated between 25.5 and 55.1% (World Health Organization [WHO], 2018a). In 2016, uncomplicated malaria was the first cause (41.7%) of outpatient consultations in public health centers and, hospitalization rate related to severe disease accounted for 20.5%, with 3.8% hospital mortality (Thomas et al., 2020). In Togo mortality rate associated with severe malaria is underestimated because most of the cases do not adhere to clinical consultations. For example, in 2017, > 60% of severe cases have not been to the hospital (Programe National de Lutte contre le Paludisme [PNLP], 2019). Children under 5 years old remain highly affected by malaria, with 35.4% uncomplicated outpatient cases, 58.4% hospitalized severe cases, and 69.7% deaths reported in 2017 (Programe National de Lutte contre le Paludisme [PNLP], 2019). Pregnant women accounted for 5% uncomplicated outpatients and 6% hospitalized severe cases (Programe National de Lutte contre le Paludisme [PNLP], 2019). To combat the disease, the country has developed and implemented several interventions based on prevention, care, and support strategies in accordance with the World Health Organization guidelines (World Health Organization [WHO], 2015; Programme National de Lutte contre le Paludisme [PNLP], 2018). However, malaria epidemiology is poorly understood in Togo, with a paucity of information on species of Plasmodium that are responsible for disease transmission. Studies on disease burden erroneously made believe that malaria cases are only related to Plasmodium falciparum (Pf) infection in Togo (Bakai et al., 2020; Thomas et al., 2020), making the standards of care unilateral with important implications in disease epidemiology.

Routine diagnosis using the standard method for the detection of Plasmodium species by examining Giemsa-stained blood smears under a microscope is the major approach being used in malaria control programmes of many tropical African settings. The method is effective and inexpensive, but it is laborious and time consuming. In addition, its sensitivity is doubtful with low parasitemia during low-transmission periods and also with the shortage of experienced technicians (Agabani et al., 1994). Polymerase chain reaction (PCR) is one of the highly sensitive methods, which has been widely used for molecular detection and identification of malaria parasites. Its effectiveness is unquestionable as it had been successfully used to detect cases of mixed infections or low parasitemia (Snounou et al., 1993c; Zhou et al., 2014). A range of PCR-based assays has been described for malaria detection including species of Plasmodium that regularly infect humans [Pf, Plasmodium vivax (Pv), Plasmodium ovale (Po), Plasmodium malariae (Pm), and Plasmodium knowlesi (Pk)] (Snounou et al., 1993a,b; Singh et al., 2004; Chen et al., 2017). In addition, nested PCR is used for the amplification of the small subunit ribosomal RNA (ssrRNA) gene of Plasmodium species (Snounou and Singh, 2002).

In order to advance knowledge of diagnosis efficacy and species of malaria parasites in Togo, this study carried out a molecular investigation on malaria clinical samples in comparison with light microscopic examination and assessed the phylogeny of Plasmodium species from southern Togo.

A total of 810 blood specimens were collected in this study. Following the diagnosis results (hereinafter referred to as test results) with Giemsa-stained thick blood smear microscopic examination, we obtained 611 positive malaria cases [referred to as microscopy-positive group (MPG)] including 610 cases of Pf and one case of Pm, while the remaining 199 cases were tested malaria negative and were referred to as the microscopy-negative group (MNG). The use of Plasmodium genus-specific nested PCR identified 565 malaria-positive cases including 536 and 29 cases from the MPG and MNG groups, respectively. In other words, nested PCR confirmed 536 out of 611 (87%) cases that were true positive from the MPG and identified 29/199 (14.6%) malaria cases that were found negative (Table 2) in the MNG. Such a finding reveals that there were 75/245 false-positive cases (30.6% error in species diagnosis) in the MPG and 29/565 (5.1%) undiagnosed cases in the MNG. This was reflected in the sensitivity and specificity values obtained (94.9 and 69.4%, respectively), suggesting that the test missed 5% of the people who had malaria and 31% of the people tested positive for the disease which they did not have. Such errors in microscopic test might be related to inadequacy of precision in parasite-based diagnosis by microscopy or low parasitemia conditions. In conjunction, PPV and NPV for the population tested in this study were 87.7 and 85.4%, respectively (Table 2). Clinically speaking, this implies that among the people who tested positive, the chance that a person with a positive diagnostic test truly has the disease was 88% and among those who tested negative, the probability that a person with a negative test does not has the disease was 85%.

Further species-specific nested PCR and species confirmation through direct sequencing of the nested-PCR products identified higher infection rate [95.5% (540/565)] for Pf than those for other species, which were 0.5% (3/565) for Po and 0.4% (2/565) for Pm. In addition, 20 cases were coinfection cases of Pf and Po [2.7% (15/565)] and Pf and Pm [0.9% (5/565)] (Table 3). However, there was no malaria case of Pv or Pk found in this study (Figure 2).

Phylogenetic relationships of unique sequences that were amplified were constructed on the basis of similarities with published homologous sequences from different countries based on neighbor-joining analysis of the 18S rRNA locus. Phylogenetic analysis of the fragments of 18S SSu rRNA gene sequences of the sample cases tested in this study showed species specificity and clustered with homologous sequences of isolates of Plasmodium species from different countries (Figure 3). Gene sequences of malaria samples that were used for the phylogenetic analysis were deposited in the National Center for Biotechnology Information (NCBI), under GenBank accession numbers [MW490726, MW492389, MW492390, MW492391, and MW504628 (Pf); MW492393 and MW492394 (Po); MW492392 and MW504627 (Pm)].

Togo, an integral part of tropical Africa where malaria is endemic, has its entire population at risk of malaria infection and disease transmission occurs all year round (World Health Organization [WHO], 2018a; Kassegne et al., 2020, 2021). Sustainable efforts have been engaged to reduce malaria burden in Togo; however, effective disease control toward elimination requires novel and innovative approaches among which diagnostics, monitoring, and surveillance are of capital importance (Chen et al., 2012; Kassegne et al., 2016, 2019).

Microscopic analysis is currently being used as the appropriate test for routine clinical diagnosis of malaria, especially in low-income settings. However, the sensitivity of the detection of malaria parasites by microscopy is approximately 50–200 parasites/μl of blood (Mbanefo and Kumar, 2020). Molecular tests are very sensitive to detect Plasmodium species, and a number of studies demonstrated that PCR is more reliable than microscopy in detecting malaria parasites in infected individuals, especially in those with low parasite density (Roper et al., 1996; Singh et al., 1996; Zhou et al., 2014). While the population tested in this study only focused on the southern Togo, the findings revealed disease prevalence (69.8%) higher than that reported (25.5–55.1%) for the entire country (World Health Organization [WHO], 2018a). Microscopic examination had 94.9% sensitivity and 69.4% specificity, showing that the test missed 120 of the people who had malaria, and one-third of the people tested positive for the disease, which they did not have, respectively. Similar observations of misdiagnosed malaria cases by microscopy have also been reported from a molecular survey of febrile cases in malaria-endemic areas along the China–Myanmar border (Zhou et al., 2014). In this study, the test showed sensitivity aligning onto that generally estimated (95%), whereas its specificity was relatively lower (69.4%) than expected (98%) as reviewed in a recent study (Mbanefo and Kumar, 2020). That is to say, the microscopic diagnosis for this study population tested people false positive for the disease they did not have. These misdiagnoses could be attributed to lack of precision and expertise in identifying species of malaria parasites, low parasite density conditions, or inadequate staining/poorly maintained microscopes. Such an interpretation is consistent with the fact that the subset of microscopy-positive PCR-negative specimens were recorded for low parasitemia from individuals who have been under antimalarial medication. Since 2005, the Togo National Malaria Control Programme has recommended two different formulations of artemisinin-based combination therapy (ACT), artesunate–amodiaquine (ASAQ) and artemether-lumefantrine (AL), for the treatment of uncomplicated malaria as well as for the treatment of unconfirmed malaria (World Health Organization [WHO], 2018a). In other words, the same care is given irrespective of whether a falciparum or unconfirmed malaria diagnosis is made. In addition, given that routine microscopic diagnosis usually does fall short in identifying other species of Plasmodium than Pf, no specific drugs effective against ovale relapse or malariae infection have been reported being used in Togo. Thus, the potential misdiagnosis of malaria parasite species by microscopy does not affect current treatment standards but, instead, has important implications in malaria epidemiology because it would identify individuals without disease but for whom the test indicates “disease” or individuals who test positive for a disease which they do not have. This may prompt resistance to artemisinin or delay in parasite clearance as patients would have been given inappropriate treatment to their ailments. Another public health consequence would be that untreated patients could become carriers for malaria transmission in areas where disease diagnosis may be already “problematic”.

PCR-based molecular methods are able to detect malaria cases that are “truly positive” as well as those with mixed infections of Plasmodium parasites (Snounou et al., 1993b; Zhou et al., 2014). In this study, nested PCR was very specific and helped detect different monospecies of malaria cases or cases of mixed infections that were not diagnosed by microscopy. Among the five known species of Plasmodium that infect humans, Pf infection (and to a lesser extent Po and Pm) is the essential cause of malaria burden in tropical Africa (World Health Organization [WHO], 2018b). In West Africa, both Po and Pm malaria cases have been identified in Benin, Ghana, and Senegal (Browne et al., 2000; Roucher et al., 2014; Trape et al., 2014; Doritchamou et al., 2018), as well as Po in Cote d’Ivoire and Comoros Islands (Bauffe et al., 2012), to mention a few. However, there have been no information publicly reported yet on Po and Pm from Togo. In this study, Pf was the most prevalent Plasmodium species detected (95.5%) among the sample cases, aligning onto the evidence that Pf is always highly endemic in West African countries because the region usually experiences long periods of rainy reason which enhances perennial or semiperennial Anopheles breeding sites. We, also detected low infection rates of both ovale and malariae cases (0.5 and 0.4%, respectively). The detection of these two “bashful” malaria parasites in such a Pf-endemic area is consistent with previous studies from West Africa, which reported that Po and Pm are often associated with Pf infections, especially in areas where falciparum is highly endemic (Mueller et al., 2007; Roucher et al., 2014). However, the Po and Pm infection rates found in this study were much lower than those reported from studies in other West African countries [e.g., 15.5% Po and 10.4% Pm infections in a community-based survey in Ashanti Region of Ghana (Browne et al., 2000); 9.2% Pm and 5.8% Po infections in a cohort in Beninese pregnant women (Doritchamou et al., 2018), or 2.5% Po and 12.2% Pm infections in a longitudinal study in Dielmo Village, Senegal (Roucher et al., 2014)]. The large-scale cohorts conducted over the years in such malarious areas might explain the high infection rates observed for both Po and Pm. Furthermore, there was evidence that Po and Pm are sympatric with Pf across the African continent and are frequently present as coinfections (Fancony et al., 2012). Low-level infection rate with Po or Pm seems to be common across tropical Africa in areas where malaria is endemic and often as complex mixed infections with Pf. The rate of Pf coinfections with Po or Pm was reported to be about 24% in the rainforest area of Ghana (Browne et al., 2000), while the proportion was much lower in other parts of West Africa such as Dielmo Village, Senegal (7.9% Pf/Pm and 1.1% Pf/Po mixed infections) (Roucher et al., 2014) or among Beninese pregnant women (6.6% Pf/Pm and 2.3% Pf/Po mixed infections) (Doritchamou et al., 2018). The low mixed infection rates that we observed in this study may be related to that all the samples were mainly collected in urban areas. However, we found infection rates of coinfection cases for both Po and Pm with Pf higher than the rates for single Po or Pm infection cases. This was consistent with previous findings from Ghana where it has been found that cases infected with Pf are more likely to carry Po or Pm than those who were not infected with Pf (Browne et al., 2000). Such a finding suggests evidence for interactions between Pm or possibly Po, with Pf infections (Mueller et al., 2007).

A series of comprehensive studies have investigated Pf in different parts of tropical Africa where malaria is endemic. However, morbidity associated with Po and Pm infections has received relatively little attention and was, therefore, markedly underestimated and unreported because they are generally regarded as a benign cause of malaria. Such interpretations could be attributed to the following reasons: (i) low incidence of the disease, (ii) lack or rarity of severe cases as they are regarded less important to Pf in terms of public health issues, or (iii) practical difficulties in microscopic identification as Po and Pm share morphological characteristics resembling those of Pf (young ring forms of the three species may be difficult to distinguish in thick blood films, especially during coinfections) (Roucher et al., 2014). Thus, in malaria-endemic areas of tropical Africa including Togo, most of all malaria attacks are erroneously attributed to Pf, probably due to underdiagnoses of Po and Pm clinical attacks. This study showed evidence of Po and Pm coexistence with Pf in southern Togo. More so, ovale or malariae malaria would have already been diagnosed by microscopy as falciparum malaria due to similarities in symptoms and species morphology. PCR-based diagnostic tests of clinical cases are therefore required for accurate species discrimination and detection.

Although considered “mild,” Po and Pm may constitute an important cause of morbidity and high risk of mortality. For example, fever episodes related to parasitemia peaks of Pm (in Liberia and The Gambia) or Po (in Senegal), in both children and adults have been reported (Miller, 1958; Greenwood et al., 1987; Faye et al., 2002). Also, Pm can cause chronic infections that can persist in the host for many years and reoccur after living in endemic areas (Siala et al., 2005); e.g., a chronic nephrotic syndrome that, once established, does not respond to treatment and carries a high rate of mortality (Mueller et al., 2007). Po shares peculiarity with Pv to form the latent stage in the liver—hypnozoite, which causes late relapse of the parasite, with new febrile episodes (after months or years) without recent exposures (Coldren et al., 2007). The reinforcement of malaria control policies for accurate diagnosis of parasites species in Togo should be, therefore, prioritized, to ensure deployment of appropriate treatment regimen for malaria patients and to provide a better knowledge on epidemiological assessments to guide control interventions.

This study publicly reports, for the first time, a molecular survey of malaria infections in Togo and also reveals the presence of other species of malaria parasites—Po and Pm, which were not previously reported. In addition, findings from this study indicated errors in microscopic examination including error in species diagnosis and undiagnosed cases. For example, in this study population, one-third of the people were microscopically tested positive for the disease they did not have. Thus, overlaps of different species of malaria could further aggravate the disease burden in Togo if appropriate actions are not taken, especially in situations where diagnosis of monospecies of malaria cases seems already “problematic.” Collectively, this preliminary study advanced our understanding of the malaria epidemiology in southern Togo and provided information to improve disease control/surveillance policy. However, large-scale studies across the whole country are required in the future to assess malariae and ovale malaria epidemiology.

The datasets presented in this study can be found in online repositories. The names of the repository/repositories and accession number(s) can be found below: https://www.ncbi.nlm.nih.gov/, MW490726, MW492389, MW492390, MW492391, MW504628, MW492393, MW492394, MW492392, and MW504627.

Permission was obtained from all malaria subjects before collecting the specimens. Blood collection was made under a study protocol reviewed and approved by the Togo Ministry of Health’s Bioethics Committee (Authorization N°019/2019/MSHP/CBRS), following institutional ethical guidelines by the Ethics Committee at the National Institute of Parasitic Diseases, Chinese Center for Disease Control and Prevention, as reported previously (Kassegne et al., 2020).

KK and J-HC conceptualized the study. KK, KA, KN, and KKK collected and analyzed the specimens. KK, S-WF, and KKK conducted the experiments. KK interpreted the data and wrote the manuscript. EMA, X-KG, and X-NZ revised the manuscript critically for intellectual content. All authors contributed to the article and approved the submitted version.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.