Background
Malaria is a vector-borne infectious disease that continues to have high morbidity and mortality globally [
1]. The primary clinical presentation of malaria is fever or flu-like symptoms or a history of fever and flu-like symptoms. Diagnosis based only on clinical symptoms has very low specificity [
2] as there is no combination of symptoms that reliably distinguishes malaria from other causes of fever or flu.
Light microscopy, malaria nucleic acid amplification (PCR) test and malaria rapid diagnostic tests (RDTs) are used for parasitological diagnosis of malaria. Malaria PCR is not commonly used due to its high cost; RDTs are now more common, but not yet the regular test in non-endemic areas; microscopic examination of stained blood films remains the standard and most commonly used diagnosis method. Although this is the standard method, it is labour intensive and requires a high level of expertise to scan thick blood films for the presence of malaria parasites and thin blood films to determine the type of malaria [
3,
4].
Malaria is prevalent in parts of Southeast Asia, South Asia, Africa and South America and, therefore, there is a demand for malaria screening in these areas [
5‐
7]. Haematology analysis technology has improved considerably in the last 70 years, particularly with the introduction of automated haematology analyzers. Modern analyzers are capable of processing hundreds of samples per hour by using flow cytometry, and such techniques could aid in the diagnosis of malaria [
8‐
11]. A recent study introduced a novel analyzer, which was able to detect infected red blood cells (iRBCs) in blood samples from mice infected with rodent malarial parasites [
12] and determine the developmental stage of cultured
Plasmodium falciparum [
13]. Although this research is promising, most analyzers using numeric and graphical data to detect malaria fail to move beyond the research phase and are not clinically applicable due to the careful monitoring required to detect malaria parasites. A more reliable and accessible detection method for malaria is a dedicated alert message/flag that is incorporated into routine complete blood count (CBC) analysis when a positive case is detected. This could support earlier detection and potentially reduce adverse outcomes related to malaria infection.
For clinicians without extensive technical experience or expertise, the BC-6800 automated haematology analyzer (Mindray, Shenzhen, China) could be used to detect malaria parasites and to calculate parasitaemia in blood samples through cytometry analysis. The BC-6800 analyzer provides CBC parameters, reticulocyte and its fractions, and nucleated red blood cells (NRBC) value. It also provide a dedicated flag for ‘InR’ and ‘InR#/InR‰’ parameters. These are research use only parameters roughly equivalent to the number of InR (InR#)/the permillage of InR (InR‰) for the malaria iRBCs in a sample [
14]. BC-6800 light scatter and fluorescence three-dimensional analysis technology (SF Cube) detects ‘iRBCs’ using signals generated by side-scattered light (SC, representing the internal cell structure and its contents), forward-scattered light (FS, indicating iRBC size), and side fluorescent light (FL, corresponding to DNA content). The flag is generated without the use of any special reagents.
The present study aimed to evaluate the utility of the infected RBC flag and parameters InR#/InR‰ in routine blood testing as a malaria screening tool in endemic areas. Furthermore, a comparison study was carried out for the flag information of CBC parameters between control group and the malaria group with P. vivax and P. falciparum.
Methods
Samples and technical principles
EDTA anti-coagulated blood samples were collected from Tengchong, Yunnan Province (an endemic area in the border region of China and Myanmar) between May and August in 2016 according to physicians’ request. Some 117 malaria-infected samples were analysed, including 96 from P. vivax-infected patients (aged 3–69 years; 80 men and 16 women) and 21 from P. falciparum-infected patients (aged 6–60 years; 14 men and 7 women). Sixty-four healthy subjects from a non-endemic area in China were collected as the control group (aged 8–70 years; 47 men and 17 women). Samples were classified as malaria positive only when microscopy examination of stained thick blood film revealed the presence of at least one of the four malaria parasite forms: ring form, trophozoite, schizont and/or gametocytes. All blood samples were examined using both light microscopy and BC-6800 haematology analyzer using the manufacturer recommended reagents, calibrator and controls. Analyzer performance was monitored daily using three levels of quality control material.
The BC-6800 haematology analyzer used sheath flow impedance, laser scatter and SF Cube analysis technology. The SF Cube analysis technology is three-dimensional using information from laser light scatter at two angles and fluorescent signals for cell differentiation and counting [
15]. In the BC-6800 differentiating (DIFF) channel, the fluorescent staining technology was adopted after the sample was mixed with DIFF lyse. For samples infected with malaria, RBC and white blood cell (WBC) sub-populations were differentiated by their size and complexity using lysing. Due to the different content of nucleic acid in WBC sub-populations, the volume of fluorescent dye staining the nucleic acid substances was different: the low-angle light scatter reflects cell size, the high-angle light scatter reflects intracellular granularity, and the intensity of fluorescent signal reflects the degree to which the cell is stained. Since multiple proteins play a role in generating increased rigidity of
Plasmodium-infected erythrocytes, that would result in resistance to lysis [
16,
17]. Through three-dimensional signal analysis of the cells processed with lyse, the DIFF channel differentiates the sub-populations, including lymphocytes, monocytes, neutrophils,
Plasmodium-infected RBC, and eosinophils, as well as identifies and flags abnormal cells such as immature granulocytes, abnormal lymphocytes and blast cells.
Comparison with light microscopy and quality control
Thin and thick blood films were prepared for staining and parasite investigation. Blood films were sent to Tengchong Laboratory in a standard slide box for staining with 3% Giemsa and microscopic examination [
18]. Parasitaemia was determined from thick blood films by counting the number of parasites per 200 WBCs. Thick blood films were classified as positive if one or more malaria parasites were observed and negative if no parasites were observed after examining at least 100 ‘oil-lens’ fields (i.e. at a magnification of ×1000). Thin smears were further examined after parasites were seen in the thick smears in order to measure parasitaemia and identify the species of malaria parasites. All the blood smears were examined using a CX21 light microscope (Olympus, Tokyo, Japan).
The malaria microscopist was blinded to BC-6800 results. Similarly, healthy controls were blinded to microscopy results. In order to assure malaria form quality of the microscopic examinations, the entire positive and 10% of the negative slides were sent to a senior malaria microscopist and re-examined at the Tengchong Laboratory. An experienced reader was assigned to re-examine the discrepant slides.
Ethical consideration
This study obtained ethical approval from the ethics committee of the General Hospital of the People’s Liberation Army. Individual informed consent was obtained from adults and from the parents or guardians of children under 18 years old when they presented in hospital. In addition, verbal assents were given to minors.
Data analysis
Data entry was performed using Microsoft Office Excel. Statistical analysis was performed using SPSS 22.0 (SPSS Inc, Chicago, IL, USA) analyzer Microsoft tool box. It was compared on the blood cell parameters and suspect flags between control group and iRBC group in BC-6800 analyzer. Pearson’s correlation or Spearman’s rank correlation was also used to determine the association between control group and malaria group (P. vivax group and P. falciparum group). Sensitivity, specificity and predictive values for the detection of different Plasmodium species were calculated and compared the P. vivax group and P. falciparum group. InR# representing the number of ‘iRBC’ was evaluated using ROC curve analysis, reporting the area under the curve (AUC) and its confidence interval (CI). ROC analysis was completed using Analyse-it v4.92.3. The Kappa value was calculated to evaluate the agreement between BC-6800 and light microscopy. A P value less than 0.05 was considered statistically significant.
Discussion
The BC-6800 haematology analyzer uses SF-Cube technology to achieve three-dimensional cell analysis for WBC, reticulocyte (RET) and NRBC with clinically relevant suspect flags [
19,
20]. It provides a dedicated flag named ‘infected RBC?’ and parameters ‘InR#/InR‰’ to represent the possible presence of
Plasmodium parasites and the causative agent of malaria infection. In this study, the parameter ‘infected RBC’ was used to analyse the diagnosis performance in patients with malaria, as well as routine parameters and suspect flags.
The diagnostic performance of the InR flag for the diagnosis of
Plasmodium was evaluated. Samples with malaria parasites show a distinct cluster and unique location on three-dimensional cell plots. The characteristics of the cluster are dependent on the species of
Plasmodium as well as the size and number of the parasite present. The sensitivity and specificity for detecting
P. vivax was not significant for
P. falciparum. This may be a result of early trophozoites of
P. falciparum that developed over 10 h in the peripheral blood concealed in the microvasculature, sinusoids or other slow blood flow before being developed into late trophozoites and schizonts [
21,
22]. Trophozoites and schizonts of
P. falciparum are rarely seen in the peripheral blood of infected patients. The ring and the gametocyte stage are the most commonly seen in a peripheral blood smear. However, the ring and the early trophozoites of
Plasmodium cannot be detected easily on BC-6800, the main reason possibly being the small size and low nucleic acid content.
InR flag being triggered by BC-6800 should meet two conditions. First, the malaria-infected specimens should be collected during the intra-erythrocytic phase
Plasmodium. Second, the number of intra-erythrocytic phase
Plasmodium should surpass a certain threshold [
23]. Some
P. vivax specimens were tested as false negatives because they did not meet these two conditions. This may be because the majority of samples among the
P. vivax group are trophozoite and ring form, which are too small to be detected easily. Both microscopy and malaria RDT are very effective at identifying ring infections. In addition to its poor sensitivity for detecting
P. falciparum, this study suggested the shortcoming should be improved in BC-6800 haematological analyzer. At present, this is the major limitation of broader use for
P. falciparum in particular. The machine could be more effective for identifying
P. falciparum in future.
Several studies report that DIFF scattergram suspect flags can provide assistance in the diagnosis of malaria in
P. vivax-infected patients [
10,
24]. Results from the present study support this finding in the iRBC group. Research shows that high eosinophil count is found in the presence of
P. vivax infection, and was observed in up to 39%
P. vivax-infected patients [
25]. In this study, results found that pseudo-eosinophilia was observed in only a small number of samples (
P. vivax 6.25% and
P. falciparum 14.29%). In
P. vivax and
P. falciparum-infected patients, results found that WBC count was higher in the DIFF channel (mild lyse) compared to the WBC/BASO channel (severe membrane destruction). A strong correlation was found between ∆WBC (WBC
DIFF–WBC
BASO) and the number of iRBC (R
2 = 0.973) for
P. vivax-infected patients (Fig.
3). A correlation was also found between parasitaemia and the number of iRBC for
P. vivax-infected patients (R
2 = 0.734; Fig.
4). This makes it possible to estimate the amount of
P. vivax after samples are tested using the BC-6800.
Anaemia a common symptom of malaria and is associated with dyserythropoiesis and ineffective erythropoiesis [
26]. The BC-6800 haematology analyzer provides additional parameters, including the immature platelet fraction (IPF) and immature reticulocyte fraction (IRF), both important in clinical markers for thrombocytopaenia and anaemia [
27‐
29]. Results from the present study show
P. vivax or
P. falciparum parasitaemia are associated with abnormal reticulocyte parameters, such as RET scattergrams and IRF. Results also show platelet-associated flags, including PLT clump, IPF and PLT abnormal in malaria-positive samples were increased compared to the control group. These findings are in accordance with the study of Dubreuil et al. [
30].
Except for cost, malaria PCR and RDTs are obviously ahead of the BC-6800 in sensitivity. It is necessary to further improve the sensitivity, especially for the ring infections, in the apparatus. In this study, the associations were limited between the haematological parameters and the species of infection considering patient age, symptoms and prior treatment.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.