2.1. Microscopy
Despite advances in diagnostic technology in the past 20 years, microscopy is still considered the gold standard for malaria diagnosis. According to World Health Organization (WHO) since 2010, all suspected malaria cases must be confirmed with either microscopy for
Plasmodium sp. identification and parasitemia or Rapid Diagnostic Tests (RDTs) in all clinical settings; nowadays, the reference method to diagnose malaria by microscopy consists of Giemsa stained thin and thick blood smears [
16]. Two slides of each type must be performed to increase diagnostic yield and all cases identified by laboratory-confirmed diagnosis should be reported to the State Health Department. Microscopy should be performed instantly, and results should be available as soon as possible and no later than 2 hours from the sampling, and in any case ≤24 hours of the patient’s presentation [
17]. In case of an initial negative result at microscopy, blood smears should be repeated at each febrile attack every 12-24 hours for a total of three sets before the diagnosis of malaria can be excluded [
18]. Thick blood films are mostly used to detect the presence of malaria parasites and to assess the parasitemia, while thin blood films are useful to identify the
Plasmodium sp. and the circulating stages of the parasite’s life cycle stages within the blood of the patient [
10]. Especially in areas where infection is not endemic and where malaria cases occur with low parasitemia, the parasite number can also be assessed by analyzing at microscopy a well-stained thin blood film as a percentage of infected Red Blood Cells (RBC) [
19]. Advantages of light microscopy include (a) low direct costs (2,5-5 Euros, excluding the cost for the microscope) in a high-volume sample; (b) good sensitivity and results in 2 hours; (c) identification of
Plasmodium sp. and stage differentiation; (d) parasitemia count; (e) drug-induced morphological changes observation; (f) the absence of parasites to assess the clearance of the plasmodia; (g) screening for other related blood abnormalities and other blood parasites (i.e.,
Babesia,
Trypanosoma,
Filaria) at once [
16]. Parasitemia is essential for the classification of malaria severity and prognosis: parasite density more than 5% is a criterion to identify severe malaria cases and parasite density counting should be continued until parasites are cleared as a follow-up to evaluate the response to the anti-malaria treatment [
18].
Unfortunately, microscopy has several limitations: it could not allow to differentiate the morphology of mature developing stages (trophozoites, schizonts, and gametocytes) between
P. knowlesi and
P. malariae, and the early ring trophozoites stage between
P. knowlesi and
P. falciparum [
20]; many parasites could be missed during the staining process causing both a reduced sensitivity of the method and an incorrect count of parasite density leading to a microscopic threshold of 50 parasites/µl; the difficulty in diagnosing of mixed infections; a sufficient parasite density and the availability of experienced laboratory personnel, particularly microscopists. The ring stages of
Plasmodium parasites might be confused with the same stages of another protozoan parasite of the red blood cells,
Babesia, or conversely, by an untrained examiner, often causing misdiagnosis [
21].
Microscopy is the only diagnostic tool able to demonstrate the presence of an active infection based on the stage identification, in fact considering the life cycle of human
Plasmodia parasites it can indicate “live” parasites. For this, microscopy still is considered the gold standard method for the laboratory diagnosis malaria (
Figure 1), although currently instruments able to automatically analyze the blood of the patients have been developed and tested [
22]. Moreover, automatic slide reading instruments or vision-based devices have been also developed and tested, even if microscopy by trained personnel is still the preferred approach [
22]. In a study conducted in 2012 an automated malaria slide scanning system, the World Health Technology (WHT) autoanalyzer, was one of first system tested performing at a level comparable to many human slide readers [
23]. Cella Vision DM96 is another digital system that applied on blood films using its Advanced Red Blood Cell Application (ARBCA) is able to recognize and classify the cell morphology of both leukocytes and erythrocytes including parasitized erythrocytes [
24].
The microscopic examination of
Plasmodium parasites using the Parasites Fluorescent labeling is another diagnostic procedure applied to the diagnosis of malaria: acridine orange staining is incubated with the patient’s blood and the DNA/RNA of the different stages of
Plasmodium sp. is marked in green and orange, respectively [
25]. Fluorescent parasites are successively detected by a conventional fluorescence microscope. The advantages of this approach are field applications because of a reduced energy requirement, stronger brightness, and contained costs. This approach allows better results compared to the Giemsa staining under a revised acridine orange staining protocol [
26]. Even if this is a feasible method which leads to fast diagnostic results in less than 1 h, trained personnel are needed to correctly label the patient’s blood samples and to correctly perform the analysis by fluorescent microscope [
27]. However, the result of acridine orange staining must be confirmed with Giemsa staining.
2.2. Rapid Diagnostic Tests
In establishing prompt malaria diagnosis, multiple Rapid Diagnostic Tests (RDTs) have been developed as a complementary test, providing a result within 15 minutes, and requiring minimal training. According to WHO recommendations [
28] in areas where microscopy or other approaches are not available, antigen based RDTs can be a valid alternative to obtain a fast and easy diagnosis of malaria and for these reasons they are often adopted in health care systems to screen patients with clinically suspected malaria, followed by microscopy.
Malaria RDTs consist of a lateral flow immunochromatographic test on nitrocellulose strip which detects either species-specific or genus-specific
Plasmodium sp. antigens or a combination of both in a finger-prick blood sample allowing the diagnosis of
P. falciparum or
P. falciparum versus non-
P.
falciparum infections even if the non-
P. falciparum malaria parasites cannot be generally revealed. Different formats of RDTs are commercially available, e.g., dipsticks, cassettes, and cards being cassettes and cards easiest to use in lack of health facilities. RDTs are easy to perform and simple to interpret, not requiring equipment and they were originally suggested as kit for first diagnostic aid in travellers to endemic areas [
29,
30,
31].
Based on the results of the WHO malaria RDT Product testing Programme, since 2012, WHO recommendations indicated the selection of RDTs according to the following criteria [
28]: for the detection of
P. falciparum in all transmission settings, the panel detection score for
P. falciparum samples should be at least 75% at 200 parasites/µl; for the detection of
P. vivax in all transmission settings, the panel detection score against
P. vivax samples should be at least 75% at 200 parasites/µl and false positive rates should be less than 10% and invalid rates less than 5% on the whole.
Antigens commonly detected in the commercially available RDTs are: 1.
P. falciparum-specific antigen Histidine-Rich Protein 2 (HRP2); 2. a pan-
plasmodium Lactate Dehydrogenase (LDH) (pan-pLDH); 3.
P. falciparum-specific LDH (
PfLDH); 4.
P. vivax-specific LDH (
PvLDH); 5. aldolase, which is also a pan-
plasmodium antigen [
32]. In
Table 1 some commercially available RDTs and the parasite species detected are reported.
HRP2 is a protein produced only by
P. falciparum, mainly by asexual stages and gametocytes, and RDTs based on it allow the benefit of
Pf specificity together with a high sensitivity. HRP2 is used in over 80% of all RDTs and to this it is commonly chosen in Africa, where 99.7% of
P. falciparum malaria cases occur [
28]. RDT-
PfHRP2 has a 95% sensitivity and a 95.2% specificity [
32]; however, at low parasitemia level (<1000 parasites/µl), the result can be interpreted as false negative due to a weak signal on the reaction’s line [
32]. False negatives can also occur with gene deletions of HRP2, and this represents a limitation in the use of HRP2 based RDTs as tests of cure due to persistent antigenemia [
32]. Moreover, false-positive results due to a cross-reaction with rheumatoid factor were rarely reported in the past [
14].
All species of malaria parasites can be detected by the pLDH assays developed with the PpanLDH or more specifically with
PfLDH or
PvLDH and in such cases most of the limitations related to gene deletions or prozone seen with HRP2 can be avoided [
32]. Furthermore, pLDH is much more effective as a test of cure having a specificity of 87% after treatment improving to 92-100% between days 7-42 [
43]. PpanLDH has also proved be able to identify
P. knowlesi with a 97% sensitivity at parasitemia >1000 parasites/µl, but only 25% when parasitemia is <1000 parasites/µl [
32]. Overall,
Pf-pLDH showed a 93.2% sensitivity and 98.5% specificity. It was demonstrated in
P. vivax, that pan-pLDH versus
Pv-pLDH has no difference and high sensitivity (>99%).
Aldolase-detecting RDTs still give a low sensitivity (80-81.4%) and they are based on this enzyme found in the glycolytic pathway of all species of malaria parasites [
44].
The main limitations of RDTs together with the risk of false positive and false negative results, include their inability to quantify the parasitemia, to distinguish among the parasitic stages and potential missing of double infections [
14].
RDTs also have potential disadvantages: for the
PfHRP2-based RDTs is of interest the inability to allow to distinguish new infections from those effectively treated and those recently acquired, related to the
PfHRP2 persistence in the blood for 1-5 weeks after an effective therapy; poor sensitivity in
P. malariae and
P. ovale detection and the heterogeneous quality of commercially available products producing the existence of batch-to-batch variation [
31]. Another weakness of the RDTs is the positive results in non-malaria febrile patients [
45]. A newly developed Highly Sensitive RDT (HS-RDT) represents a promising tool to better detect
Plasmodium species in the blood of infected subjects [
46].
The United States Food and Drug Administration has approved only one RDT (BinaxNow
TM), a card combining HRP2/Aldolase with a 95.3% sensitivity, and 94.2% specificity for
P. falciparum [
28] and 68.9-74.6%, and 99.8% for
P. vivax, respectively. BinaxNOW
TM RDT has a
P. falciparum line linked to HRP2 (T1), and a pan-malaria line (
Pv,
Po, or
Pm) linked to aldolase T2 (
Figure 2). When the result is the appearance of both T1 and T2 lines, it cannot be used alone to distinguish whether this is the case of a multi-species infection involving
P. falciparum mixed with a non-
falciparum species or the case of a high
Pf parasitemia because the aldolase is a preserved enzyme in all species of malaria parasites, including
P. falciparum.
The originally proposed use of RDTs was as tool for self-diagnosis in high-risk groups, especially travellers in malaria endemic areas after appropriate training allowing timely an adequate management and avoiding over-diagnosis of malaria on-site and inappropriate antimalaria treatment. This use of RDTs is still controversial, although recent studies have produced encouraging results.
Figure 2.
Immunocromatographic assay for the detection of Plasmodium sp. antigens in blood samples: P. falciparum (Pf), P. malariae (Pm), P. vivax (Pv), and P. ovale (Po). C is the control line for human blood, T1 line specific to P. falciparum (Pf) Histidine-Rich Protein2 (HRP2), and T2 line is for the parasite lactate aldolase. A Pf or mixed infection on the left and a Pf infection on the right [modified form14].
Figure 2.
Immunocromatographic assay for the detection of Plasmodium sp. antigens in blood samples: P. falciparum (Pf), P. malariae (Pm), P. vivax (Pv), and P. ovale (Po). C is the control line for human blood, T1 line specific to P. falciparum (Pf) Histidine-Rich Protein2 (HRP2), and T2 line is for the parasite lactate aldolase. A Pf or mixed infection on the left and a Pf infection on the right [modified form14].
2.5. Molecular Methods
According to the reasons described above, microscopy still remains the reference method for the laboratory diagnosis of malaria, while RDTs represent an important diagnostic aid over more traditional methods and molecular methods are currently used as confirmatory assays. In fact, molecular methods are crucial when morphological characteristics of the parasites overlap each other, or parasite morphology is altered by drug treatment, in case of mixed infections by different
Plasmodium species, incorrect storage of the samples, or when sub-microscopic parasitemia occurs [
14,
21,
71].
Overall, Nucleic Acid Amplification Tests (NAATs) are at least 10-fold more sensitive compared to microscopy having a detection limit of about 0.2-6 parasites/µl of blood, based on the assay and the
Plasmodium sp. involved [
14]. The overall category of NAATs used to detect different
Plasmodium sp. in the blood includes PCR (nested-PCR, multiplex-PCR, real-time PCR), Loop-Mediated Isothermal Amplification (LAMP), molecular-based Point of Care Test (POCT), Nucleic Acid Sequence-based Amplification, Rolling Circle Amplification, Recombinase Polymerase Amplification (RPA), and Clustered Regularly Interspaced Short Palindromic Repeats (CRISPR) [
21].
The gold standard among the DNA detection methods to diagnose malaria is widely considered to be the nested-PCR described by Snounou et al. [
72,
73] targeting the 18S-rRNA gene, including a genus-specific characterized sequence of about 1.2 Kb containing all the
Plasmodium human-infecting species-specific sequences. A modified nested-PCR assay was developed to improve the original method [
74].
Newly developed NAATs include additional target genes, such as mitochondrial DNA (mtDNA), highly sensitive because of the large number of target copies (≈20 copies) [
75] and allowing the detection all human
Plasmodium sp. together with the 18S-rRNA, and other targets, such as
P. falciparum stevor multigene family [
76], telomere-associated repetitive element [
77], and
P. vivax Pvr64 sequence [
78].
Several nested-PCR (nPCR) or semi-nested Multiplex-PCR (SnM-PCR) are reported in the literature [
21]. Small subunit ribosomal RNA (ssRNA) genes are the most used targets of primers used in most of the nested-PCR based assays as such genes are widely used for phylogenetic analysis and are also well characterized from various
Plasmodium sp. [
14]; for the same reasons the highly conserved dpfk13, encoding the Kelch13 protein are also used as target [
79]. When at microscopy morphological problems hinder the identification of malaria parasites at the species level nested- and multiplex-PCR results can give an accurate identification [
80]. However, these techniques present limitations in their use in low-resource settings or at point-of-need and have some disadvantages: are expensive, time-consuming, require a reliable power supply, require time for the samples preparation, reaction set-up (storage of the reagents, separate areas of work to prevent contamination), time to the end of the reaction, and the analysis of the results.
Several Real-time Polymerase Chain Reaction (Real-time PCR) assays to detect and identify the different
Plasmodium sp. (
Figure 3) in a single reaction have been developed to resolve most of the difficulties related to the use of the nested-PCR assays [
13,
21,
27,
81,
82,
83,
84].
Real-time PCR is cost-effective, with high sensitivity and specificity, although it is not included among the rapid methods for the initial diagnosis of malaria, it is fast and requires about 1.5 hours [
14]. Real-time PCR assays are potentially able to detect both low parasitemia levels and mixed infections [
85], and for this reason they should be applied not only for the diagnosis but also to prevent of drug-resistant strains to emerge as consequence of misdiagnosis with other methods and related mistreatment and for quality control purposes [
86].
The development of commercially available DNA Loop-mediated Isothermal Amplification (LAMP) - based assays is one of the most recent evolutions of DNA amplification assays for the laboratory diagnosis of malaria. It is a simple method based on the isothermal amplification, not requiring special equipment, and producing results that can be read visually or with a Real-time turbidimetry. LAMP allows to reduce the time for the result within the recommended 2 hours for the diagnosis of malaria and showed to have high efficiency, allowing DNA amplified 10
9-10
10 times in 15-60 min [
87,
88,
89]. Advantages of LAMP based assays also include the use of small amounts of blood sample on filters papers and the tolerance to inhibitory substances present in blood samples (hemoglobin/immunoglobulin), but it currently lacks sufficient accuracy [
21]. LAMP based assays might be an alternative to the other PCR methods useful particularly in remote areas because the reaction can simply run in one tube at a constant temperature, not requiring a thermal cycler and producing a rapid malaria diagnosis [
90]. Real-time PCR and LAMP assays allow results within a clinically relevant time frame, but they have the same disadvantage common to all NAATs: a positive result can indicate either current or recent past infection and cannot be used to differentiate among these two conditions.
Failure in diagnosing malaria with a PCR-based methos can occur when parasites have genetic diversity in the sequences of the target sequence of the primers or when the target gene is present at a very low copy number causing a lower amplification efficiency and consequently a reduced sensitivity [
91]. Several quantitative PCR (qPCR) assays have been described to successfully detect
Plasmodium parasites both in clinical settings and in asymptomatic subjects [
92].
It cannot be ignored that molecular assays can detect the parasitic DNA while they are not able to distinguish in the blood sample among DNA derived from live parasites, residual DNA derived from destroyed asexual stage or circulating gametocytes which can persist after a successful therapy in submicroscopic quantities causing the permanence of DNA up to weeks after a malaria resolved episode, and consequently a risk of false positive results producing the recurrence to unnecessary anti-malaria treatment. However, in experimental conditions the clearance of parasitic DNA from the blood in an animal model was demonstrated within 48 hours after malaricides treatment; it can be inferred that detecting Plasmodium DNA in a blood sample belonging to a subject with clinical suspicion of malaria could be a sign of active infection, albeit no parasites are revealed at microscopy in the same sample and the result of the NAAT stimulate to repeat blood sampling from the subject to exclude malaria as good practice.
Molecular assays are not indicated for monitoring anti-malaria treatment because in the case of recent or treated infections they can remain positive for up to four weeks (depending on the starting parasitemia), even in absence of viable parasites. Different molecular assays for the diagnosis of malaria, often developed in-house, are widely spread particularly in non-endemic areas stimulating in 2008 the establishment of the WHO International Standard for
Plasmodium falciparum DNA for NAT-based assays whose use is recommended for the quality control of the reaction and the assessment of the analytical sensitivity of different assays allowing comparative evaluation among their results [
93].
Recombinase Polymerase Amplification (RPA) allows the amplification of single-stranded DNA, double-stranded DNA, methylated DNA, and miRNA [
94]. The RPA reaction starts when a recombinase-primers complex is created by the binding of a recombinase protein to primers in presence of ATP and high molecular polyethylene glycol. The combination of the isothermal RPA with the lateral flow detection is an approach to improve molecular diagnostic tools for
P. falciparum identification in resource-limited conditions. The system requires none or little instrumentation for the reaction as the result can be read-out with the naked eye. The method was demonstrated to be highly sensitive showing a detection limit of 100 fg and 500 fg, respectively, corresponding to approximately four and 20 parasites/reaction [
95]. RPA reaction allows multiplexing highly depending on target sequences, amplicon size, and primer design [
94]. Different detection techniques can also be combined with RPA detection: bridge flocculation assay [
21], gel agarose, colorimetric fluorescence [
95], quantum dots [
96], electrochemical [
97] and surface-enhanced Raman scattering detection [
98], and for the end-point detection in malaria diagnosis the application of SYBR Green I was also described [
99].
The most used in-house and commercially currently available molecular assays are reported in
Table 3.
2.5.1. Molecular-Based Point of Care Test
The use of Point of care test (POCTs) is spread in remote areas having insufficient laboratory infrastructures and not routinely used in malaria non-endemic areas. The POCT should be equipment-free, and delivered, sensitive, specific, user-friendly, rapid, affordable, and robust [
21]. Numerous studies have been described the use of nucleic acid testing based-POCT for the detection of
Plasmodium sp., but a commercial product is not yet available due to technical obstacles such as availability of dedicated thermocycler, optimization of each reaction with suitable materials, and handling of NAATs specific reagent. Isothermal amplification techniques such as LAMP and RPA represent now the most promising techniques to be deeper tested as molecular-based POCTs candidates for the laboratory diagnosis of malaria since they require a simple instrument, using reduced energy and time to achieve a sensitive target detection [
111]. Another potential method suitable to be a POCT for the diagnosis of malaria is the microfluidic, either conventional or paper-based assay, which can overcome most of the obstacles in sample preparation, adequate amplification, and detection of genomic targets [
21].
2.6. Innovative Recently Developed Methods
Recently developed promising methods such as droplet digital PCR (ddPCR) and Next Generation Sequencing (NGS) were proposed to be used in different fields of malaria investigation including basic research and diagnostic purposes. ddPCR is a digital PCR method using a water-oil based emulsion technology providing absolute and direct quantification of DNA target not requiring a standard curve [
112]. ddPCR provides an accurate and absolute quantification by counting the DNA molecules encapsulated in approximately 15,000 discrete, volumetrically defined, water-in-oil droplet partitions that are submitted to endpoint PCR [
112]. These techniques were described in laboratory setting in the detection of almost all
Plasmodium species showing better sensitivity than qPCR.
NGS is a sophisticated method applied to better understand malaria transmission pattern and investigate the malaria parasites movement [
113] and for the identification of multidrug resistance related genes in
Plasmodium species as the major therapeutic barriers nowadays recognized [
114].
The miniature direct-on-blood PCR nucleic acid lateral flow immunoassay (mini-dbPCR-NALFIA) is a newly developed and easy-to-use molecular assay proposed for the laboratory diagnosis of malaria in resource-limited settings [
115]. Compared to traditional molecular methods, mini-dbPCR-NALFIA is innovative as it does not require DNA extraction and is based on the use of a handheld, portable thermal cycler able to run on a solar-charged power pack or incorporated as a miniature thermal-cycler making the assay well-adapted to resource-limited settings. In addition, for the result read-out a rapid lateral flow strip is used enabling the differentiation of
Plasmodium falciparum and non-
falciparum infections.
More recently, an inception-based capsule network was described as innovative approach to distinguish parasitized and uninfected cells from the analysis of microscopic images [
116]. This diagnostic model incorporates neural networks based on Inception and Imperative Capsule networks operating the detection of malaria parasites in microscopic images by classifying them into parasitized and healthy cells. The proposed system is more accurate and faster compared to traditional manual microscopy with an accuracy of 98.10% on the test, while on the 20% split, it achieves an accuracy of 99.3%. These experimental results are encouraging, and the developed model is robust and flexible and has outperformed competing models [
116].
“Digital diagnosis” includes the various aspects of digitalization, such as automation in the visualization/analysis of the data deriving from microscopy, RDTs results, and analysis of electronic health records/clinical symptoms using web-based/mobile phone applications, which could enable a graphical user interface and an ease access [
117].
Matrix-assisted laser desorption/ionization - Time of Flight Mass Spectrometry (MALDI-TOF MS) has brought a revolution in the diagnostic practice for the identification of bacteria and is widely recognized as a method of analysis fast, and robust, inexpensive, with minimal risk of operator bias. Stauning et al. reported a potential application of MALDI-TOD MS providing proof-of-concept for MALDI-TOF MS-based diagnosis of human malaria [
118]. The study concluded that MALDI-TOF MS can be applied to the detection and quantification of
P. falciparum in human blood albeit not yet applicable to diagnostic practice. Studies on clinical samples together with the development of novel sample processing protocols are required to further develop the method before considering its application to the laboratory diagnosis of malaria.
Recently, Obeng-Aboagye et al. [
119] demonstrated that levels of pro-inflammatory cytokines can be used as potential biomarkers for severe malaria, correlating with disease severity. IL-1 βand IL-17A showed good diagnostic potentials and can be considered for use in clinical practice to target treatment.
In
Figure 4, the milestones of the introduction of diagnostic assays for malaria through the years is reported.