Background
Malaria remains a serious threat to public health around the world, with an estimated 228 million cases and 405,000 deaths in 2018 worldwide [
1]. The main epidemic areas of malaria are distributed in Africa (93%), followed by Southeast Asia (SE Asia) (3.4%) and the Eastern Mediterranean Region (2.1%) [
1]. The World Health Organization (WHO) hopes to eliminate malaria in at least 35 additional countries (based on data from 2015) by 2030 [
2]. To assist in this goal, China planned to eliminate the disease by 2020 [
3]. In China, the occurrence and burden of indigenous malaria have rapidly decreased due to an integrated malaria control and elimination strategy implemented in 2000 [
4]. It is worth noting that as of 2017, no cases of indigenous malaria have been reported in the entire country. In accordance with the nationwide status, the control of indigenous malaria in the Hubei Province of China was successful, and zero indigenous malaria cases have been reported in Hubei since 2013 [
5]. However, imported malaria has gradually become a major threat in need of a process of prevention, control and elimination in endemic and non-endemic areas globally, including in China. Furthermore, it is worth mentioning that imported malaria cases from Africa and SE Asia have been gradually increasing yearly in China, including in the provincial capital of the Hubei province, Wuhan [
4,
5]. For efficient control and in order to achieve malaria eradication before 2020 in China according to the state plan [
3], imported malaria patients in China, especially in first-tier cities with a high population density and mobility, such as Wuhan, need to be quickly noticed.
There are five primary
Plasmodium species causing malaria in humans, including
Plasmodium falciparum,
Plasmodium vivax,
Plasmodium ovale,
Plasmodium malariae, and
Plasmodium knowlesi. Different
Plasmodium species infections can cause dissimilar disease characteristics. Amongst them,
P. falciparum is responsible for most of the morbidity and mortality of humans in sub-Saharan Africa, and other malarious tropical areas around the world [
6,
7]. Populations such as young children, older adults, immunosuppressed patients and travellers from malaria-free zones to epidemic areas are particularly susceptible to falciparum malaria.
Plasmodium falciparum can also increase the risk of severe malaria and adverse complications in pregnant women, such as maternal death, miscarriage, stillbirth and neonatal death. If patients fail to get diagnosis and treated within 24 h after the onset of falciparum malaria, it may be fatal [
8].
Plasmodium vivax is generally the culprit for the majority of malaria cases in areas with tropical and temperate climates and is considered to be a neglected tropical disease [
9,
10]. Severe
P. vivax malaria cases have been reported in endemic areas of (sub) tropical countries. For
P. vivax, a dormant form named hypnozoite, could stay in the liver and may lead to the malaria relapses several months or even years later [
8].
Plasmodium ovale is mainly prevalent in Africa and Asia, particularly in West Africa [
7].
Plasmodium ovale seldom leads to severe malaria among people in endemic regions, but can cause severe malaria disease in naive visitors [
11,
12]. It is comprised of two subspecies,
P. ovale curtisi and
P. ovale wallikeri [
13].
Plasmodium ovale curtisi has a significantly longer latency duration than
P. o. wallikeri [
14,
15]. Like
P. vivax,
P. ovale can also form hypnozoites in the host liver, which makes the prevention and treatment of both vivax and ovale malaria more difficult [
8].
Plasmodium malariae cases have been reported worldwide and are particularly prevalent in West Africa.
Plasmodium malariae usually causes the mildest infections and very rarely has life-threatening results. However, it may cause splenomegaly or renal damage during chronic infection [
7].
Plasmodium knowlesi was initially found in SE Asia in wild macaques, but it can also infect humans [
16]. Although human-mosquito-human transmission has not been reported for this malaria species, it is a health threat to people in forested areas of SE Asia because it may cause organ failure or even fatal outcomes [
8,
17,
18]. Beyond that, simultaneous infections with more than one human
Plasmodium species commonly happen, especially in endemic regions [
19]. However, the influence of malaria caused by
P. vivax,
P. ovale and
P. malariae as well as mixed infections is frequently underestimated [
12,
15,
20‐
22]. Most importantly, different
Plasmodium infections should be treated with the corresponding chemoprophylaxis and appropriate treatment principles [
8]. Therefore, for timely and efficient treatment and to prevent malaria transmission, early and accurate diagnosis of
Plasmodium species infection is meaningful.
As it is inexpensive and has a relatively high sensitivity for species recognition and even parasite density quantification, microscopic examination of thick and thin blood smear is traditionally regarded as the gold standard for malaria detection [
23,
24]. However, the use of microscopic examination has been restricted because it requires well-trained laboratory personnel and shows less sensitive for low parasitaemia levels [
25‐
27]. Besides, different observers may have two to three-fold discrepancies in parasite quantification [
28]. More importantly, mixed infections of different
Plasmodium species are frequently missed [
29]. The two subspecies of
P. ovale (
P. o. curtisi and
P. o. wallikeri) cannot be distinguished by microscopy. Misidentification may also happen between different species [
30]. All of the above may induce the unreasonable usage of anti-malarial drugs, hamper parasite clearance and lead to the transmission of malaria, and even anti-malarial drug resistance. The rapid diagnosis test (RDT) as an immunologic method usually targets
Plasmodium-specific antigens in blood samples such as histidine-rich protein 2 (HRP2) and lactate dehydrogenase (LDH) [
7]. RDT offers a rapid diagnosis and can be conducted by users without extensive training. It often has comparable sensitivity to microscopy. However, RDT can only distinguish falciparum and non-falciparum
Plasmodium species infection [
31]. Furthermore, false negative results by RDT were also reported in previous studies [
32‐
34]. Therefore, molecular tools such as nested PCR and real-time PCR that could not only detect
Plasmodium infection, but also allow for accurate species identification by primer design should be used. Nested PCR with two rounds of amplification can improve the detection sensitivity. Sequencing of products from nested PCR represents a further reliable guarantee for species identification [
35,
36]. The major advantages of real-time PCR include direct result reading without the need for downstream analysis or quantification of the DNA copy number [
37,
38].
Using suitable and effective methods to timely identify imported malaria is essential for successfully completing the target assessment task for malaria elimination in China, Africa and SE Asia. Therefore, the present study aimed to confirm the Plasmodium species involved in all imported malaria cases in Wuhan by using microscopy, RDT, nested PCR and real-time PCR. The performances of these methods in malaria diagnosis were assessed to supply more reference values for their future use. Analysis of their diagnostic results revealed the prevalence and characteristics of imported malaria in Wuhan, China.
Discussion
In 2020, China hopes that national malaria elimination will become a real possibility based on effective control measures [
4]. Although no indigenous malaria case has been reported in China since 2017, imported malaria is becoming a serious obstacle to malaria elimination. If China plans to achieve the stated goals on schedule, continuous surveillance of imported malaria is an essential measure. As an important city in Central China, Wuhan has a population of 11.08 million. It also faces similar problems: imported malaria cases increased annually accompanied by the disappearance of indigenous malaria cases since 2013 [
5]. Thus, it is necessary to monitor the epidemic status of imported malaria in China, including in Wuhan. Furthermore, identification of the malaria parasite species with timely and effective methods could provide valuable information for developing suitable clinical treatment.
In the current study, several discrepant results were obtained by the four different diagnostic methods, which ultimately revised 5 cases from single to mixed infection and 6 cases from negative to positive. The confirmed data revealed that 95.5% of the imported malaria patients had returned from Africa, and
P. falciparum was the dominant imported species. Analysis of the performance of the diagnostic methods involved also revealed some limitations in their usage. Of the 243
P. falciparum cases identified by microscopy, one case later proved to be infected with
P. falciparum plus
P. ovale by both molecular tests. Two microscopy-positive
P. vivax cases were also found to be cases of
P. vivax plus
P. ovale. Likewise, two patients with single
P. ovale infection by microscopy were proven as
P. ovale plus
P. falciparum infection by nested PCR. In previous studies,
P. ovale was also frequently reported to be involved in mixed infection with other
Plasmodium species [
20,
46,
47]. These results demonstrate the limitations of microscopy in the identification of mixed infection and the highly specialized requirements of the trained observers [
48]. The identification of
P. o. curtisi and
P. o. wallikeri in 20 single
P. ovale infection cases, which were unable to be distinguished by microscopy, also highlighted the advantages of molecular tools in species identification. Their diagnosis will offer a clue for the precise treatment of malaria patients. Subsequently, nested PCR also proved 6 cases as
P. falciparum infection, which included 5 microscopy-negative but RDT-positive cases and one double-negative case. This was probably because the level of parasitaemia was too low to be observed in these cases. As reported in previous studies [
49,
50], microscopy has a level of detection about 50–500 parasites/μl for
Plasmodium, but the threshold, 50 parasites/μl, is rarely noticeable unless under optimum conditions. This means that it is quite difficult for observers to accurately identify the species with relatively low parasitaemia, especially in mixed infections.
In the analysis of sensitivity, specificity, PPV and NPV, nested PCR-based detection was chosen as the reference standard. Actually, nested PCR is commonly selected as a reference method [
51,
52]. By comparing to nested PCR, microscopy (96.77%) showed similar sensitivity to real-time PCR (96.37%) for
P. falciparum identification. However, the specificity of microscopy was only 91.67%, lower than the 100% of rea1-time PCR detection. By designing complementary probes to target gene sequences, the Taqman probe is another guarantee besides the primers to enhance the specificity and avoid false-positive results [
38]. The NPV for
P. falciparum detection was approximately 85% in both microscopy and real-time PCR detection, indicating a relatively high false-negative rate of approximately 15% for these two methods. A false-negative rate as high as 19.4% for microscopy was reported in a previous study [
53]. In another study, NPV was as low as 65.8% for microscopy detection [
54]. These indicated the persistent limitations and instability of microscopy diagnosis. For real-time PCR in the present study, its high false-negative rate was mainly attributed to its failure to diagnose 9
P. falciparum cases consisting of 7 samples with low parasite density (0–500 parasites/μl) and 2 mixed infection samples. In a previous study, the same real-time PCR primers and probes were assessed [
55]. Although only 198 of 207 (95.7%) malaria samples were accurately diagnosed at the species level, the NPV for
P. falciparum detection was 100%. This indicated that diagnostic performance may vary with the reaction conditions, including the DNA quality, reaction system and even the performance of the laboratory technicians, remaining us of the significance for optimization of the reaction system. For
P. ovale, the sensitivity (88.00%) of microscopic identification was relatively low, and the NPV for
P. ovale was 98.91%. Two of the 3
P. ovale cases that failed to be detected by microscopy were proved to be
P. ovale plus
P. vivax mixed infections. This may be due to the similar morphology between
P. ovale and
P. vivax [
12]; as a result, the species with a relatively lower parasitaemia level may be hidden. The four indexes (Table
4) in real-time PCR detection were all 100% for
P. vivax,
P. malariae and
P. ovale detection, revealing the high consistency between the two molecular tools in the identification of these three species.
Plasmodium malariae could be accurately diagnosed by all three methods. This may be because only 6
P. malariae cases were involved in the study. The primers for real-time PCR used in this study have been used in several previous studies [
35,
56]. However, it has been reported that simultaneous amplification of different templates was difficult and that the primers preferred to amplify species with higher parasitaemia as a result of competition for the shared primers (Plasmo1 and Plasmo2). In the present data, 7 cases out of the total 9 nested PCR-positive patients failed to be identified by real-time PCR. Actually, all had low parasitaemia and the other two had mixed infections. Similar phenomena were also reported in previous studies [
57,
58]. A false negative result means a person with
Plasmodium infection failed to be diagnosed. This could cause a misdiagnosis, delayed treatment of malaria and even lead to a life-threatening problem, especially for
P. falciparum infection. Beyond the individual consequences, false-negative patients may also spread malaria and produce resistance to anti-malarial drugs if given incorrect treatment due to the lack of an exact diagnosis [
53].
Although nested PCR showed the highest sensitivity and real-time PCR had the best specificity, the role of microscopy and RDT is not controversial. Use of the two molecular methods relies on DNA extraction from blood samples. If errors happen during the extraction, a false-negative would probably occur. For example, in a previous study, 10 malaria cases failed to be identified by molecular methods due to problems with the nucleic acid extraction, and the reasons were only identified after repeated experiments [
59]. Therefore, the combined use of the four methods is quite necessary and highly recommended if conditions permit. Both microscopy and RDT can offer a preliminary result and suggest further diagnostic directions. Based on this valuable information, the
Plasmodium species can then be confirmed by molecular methods, including nested PCR and real-time PCR.
The origin of infection of the patients was consistent with a previous report referring to the epidemiologic features of imported malaria in China [
60]. In the epidemic regions of Africa and SE Asia, the natural environment, healthcare and situation of economic development may have primary influences on malaria infection. Chinese people overseas, especially those that do not live for long periods in epidemic areas, usually lack adaptive immunity and related awareness of local
Plasmodium infection [
60]. Additionally, there are frequently asymptomatic patients [
61]. Workers returned from these areas, particularly West Africa, Central Africa, and Southern Africa, should be encouraged to undergo malaria testing. As the dominant malaria species, the prevalence and origins of
P. falciparum were similar to other studies [
53,
62]. Therefore, prevention and treatment against
P. falciparum should be highly considered in Wuhan. Over the years (from 2014 to 2018), the
Plasmodium species leading to imported malaria were becoming diversified, just as in other provinces of China, including Shandong [
59], Zhejiang [
62], Shanxi [
63] and Chongqing municipality [
64]. In this period, the imported non-falciparum cases and
P. vivax/
P. ovale cases showed no significantly increasing trend by year, but more attention should be focused on this increasing
Plasmodium species diversity. The increase in imported
P. vivax and
P. ovale cases may cause the re-introduction of malaria into regions without cases [
65,
66].
Plasmodium malariae can survive in humans for many years and keep infectivity of its vector at a truly low parasitaemia load [
67]. The sharp decrease in imported malaria cases in 2017 may be a good sign, indicating the successful performance of malaria control. However, the cases increased again in 2018, warning us about the challenges of complete elimination of
Plasmodium infection worldwide, even in China.
Additionally, results in the present study showed no significant seasonal correlation (P = 0.759) by the Pearson correlation analysis among the imported malaria cases during the study period. One small peak of the seasonal index [
68] appeared in May (1.44) to June (1.32). Another study of malaria in Hubei province from 2005 to 2016 revealed that indigenous malaria had an obvious seasonal fluctuation during their study period, with a peak of seasonal index in August (2.26). However, only three small peaks of seasonal index (all approximately 1.3) for imported cases were found, including one in January, a second one in May and the last one in September [
5]. This indicated that there are differences in the incidence of monthly cases between indigenous and imported malaria. Comparisons with five neighbouring provinces/cities of Hubei also showed that no obvious correlation existed between season and imported malaria cases in these regions [
64,
69‐
72]. Even so, two time slots should be paid more attention for the prevention of imported malaria. One is the Chinese Spring Festival, which leads to the most frequent migration between provinces in China and brings many overseas Chinese home [
5,
63,
64,
71]. Another is the anopheles active phase [
5,
64,
70‐
72]. Importantly, commuting and the movement of imported malaria cases between provinces should also be of concern, which has been reported in another city of China [
71].
Indeed, besides an early and precise diagnosis for proper treatment, some preventative measures should also be emphasized from two major aspects before people leave for an epidemic area: (1) Prevention of mosquito bites. Travellers could protect themselves by using insect repellents, insecticide-treated nets, aerosol sprays and protective clothing. (2) Proper usage of chemoprophylactic anti-malarial drugs. Anti-malarial prophylactic regimens may not provide complete protection, but proper chemoprophylaxis could significantly lower the risk of fatal malaria. According to the prevalence of Plasmodium species at the destination and the conditions of the travellers (healthy adults, children, pregnant women or immunosuppressed travellers), different chemoprophylactic schedules should be advised by the local government according to the guide on international travel and health published by the WHO. Only a combination of prevention and treatment can lead to the achievement of the goal of malaria elimination around the world.