Platelet distribution width, mean platelet volume and haematological parameters in patients with uncomplicated plasmodium falciparum and P. vivax malaria

Background: The association between the haematological profile (including abnormal platelets) and malaria is not completely understood. There are few published data on haematological profiles of malaria patients in areas with unstable malaria transmission. The current study was conducted to investigate if the haematological parameters, including platelet indices, were reliable predictors for microscopically-diagnosed malaria infection. Methods: A case-control study with a total of 324 participants (162 in each arm) was conducted at the out-patient clinic of New Halfa hospital during the rainy and post rainy season (August 2014 through to January 2015). The cases were patients with uncomplicated Plasmodium falciparum (107; 66.9%) and P. vivax malaria (55, 34.0%) infections. The controls were aparasitemic individuals. The haematological parameters were investigated using an automated hemo-analyser. Results: There was no significant difference in the mean (±SD) age between the study groups; however, compared to the controls, patients with uncomplicated malaria had significantly lower haemoglobin, leucocyte and platelet counts, and significantly higher red cell distribution width (RDW), platelet distribution width (PDW) and mean platelet volume (MPV). Conclusions: The study revealed that among the haematological indices, PDW and MPV were the main predictors for uncomplicated P. falciparum and P . vivax malaria infection . Abbreviations: OR: odds ratio.


Introduction
In spite of the preventative measures, malaria remains a major public health concern. Malaria is responsible for 781,000 deaths in a year, the majority of which are in Sub -Saharan Africa 1 . A correct diagnosis is one of the most important tools in the management of malaria. It has been recommended that all persons with suspected malaria should have a parasitological confirmation of diagnosis 1 . Microscopic examination of malaria consists of the identification of parasite species in thin and/or thick blood films, which is the "gold standard" for malaria diagnosis 1,2 . Microscopy requires trained technicians, and well-maintained microscopes with a perfect quality management system. However, acceptable microscopy services are not widely available for the diagnosis of malaria in some areas where malaria is endemic e.g. in communities in Sub-Saharan Africa 1 .
Previously, measurement of haematological blood parameters was unreliable due to intra and inter-method variation. Nowadays, automated analysers have replaced the traditional methods. Automated analysers are available in most settings and can give reliable results within a short period of time. There is a universal trend toward using these to aid the presumptive diagnosis of malaria infection 2,3 .
Previous studies have reported different results levels of sensitivity and specificity of haematological parameters as predictors of malaria infection [4][5][6][7][8] . There is no published data on haematological changes in patients infected with malaria parasites in Sudan, where malaria in the major health problem 9 . The current study was conducted in New Halfa, eastern Sudan to investigate the haematological changes observed during malaria infection and to assess the reliability of the haematological parameters used for diagnosis.

Methods
A case-control study was conducted at the out-patient clinic of New Halfa hospital during the rainy and post rainy season (August 2014 through to January 2015). The cases were patients with symptoms and signs of uncomplicated malaria and who were confirmed to be infected with P. falciparum or P. vivax by microscopic examination of Giemsa stained blood smears during the study 10 . The controls were the patients that presented to the same clinic with symptoms of malaria but were found to have negative blood films for malaria. After the participants (or their parents/legal guardians if they were minors) provided written informed consent, a clinical history was gathered using questionnaires. Weight and height were measured and body mass index was expressed as kg/m 2 .
2ml of blood was taken from each participant and placed in a container with EDTA, and a complete hemogram was performed using an automated hematology analyser (Sysmex XN-9000; Hyogo, Japan), following the manufacturers' instructions as previously described [11][12][13] . The hemogram included measuring the haemoglobin level, leucocyte count and platelet indices, namely platelet count, mean platelet volume (MPV), and platelet distribution width (PDW).
Thick and thin blood films were prepared and stained with 10% Giemsa to microscopically confirm which participants were infected. If the slide was positive, the parasite density was measured by counting the number of asexual parasites per 200 leukocytes, and multiplied against the participants own leucocytes number/μL. The blood films were considered negative if no parasites were detected in 100 oil immersion fields of a thick blood film.

Statistical analysis
A minimum sample size of 162 participants for each arm of the study was calculated assuming that 10% of participants would have incomplete data. In this way, it would be possible to calculate a significant difference (at α = 0.05) in the means of the proposed variables -mainly haemoglobin, red cell distribution width (RDW), leucocytes, platelets counts and PDW -between the cases and the controls, at 80% power.
Statistical analysis was performed using SPSS for Windows, version 20.0 (SPSS Inc., Chicago, IL, USA). Proportions of the studied groups were expressed as percentages and compared using the chi-squared test. Continuous data were checked for normality using the Shapiro-Wilk test. The means (±SD) or median (IQR) were used to describe the studied variables, depending if they were normally or non-normally distributed. The t-test (or Mann-Whitney U test if the data were not normally distributed) evaluated the differences between the studied groups. Binary regression was calculated, where malaria was the dependent variable and medical and haematological indices were the independent variables. Diagnostic screening tests were used to determine the diagnostic cut-offs of various parameters (based on test sensitivity and specificity) using the receiver operating characteristic (ROC) curve. P < 0.05 was considered statistically significant.
Results 107 (66.9%) and 55 (34.0%) of the uncomplicated malaria cases were infected with P. falciparum and P vivax, respectively. There was no significant difference in the age or BMI between the cases and the controls. Patients had significantly higher body temperature than the controls (Table 1). Their ages ranged between 1.1−55 years Compared with the controls, patients with uncomplicated malaria had significantly lower haemoglobin levels and lower leucocyte, lymphocyte, neutrophil, and platelet counts, but significantly higher RDW, PDW and MPV (Table 2).
A receiver operating characteristic (ROC) curve was used to determine the cut-offs for haemoglobin levels, RDW, leucocytes and platelet counts, PDW and MPV for prediction of malaria infection.
The area under the ROC curve is shown in Table 3 and Figure 1, which failed to confirm predictability of hemoglobin, RDW, leucocytes and platelet count. Poor and fair predictability of PDW and MPV for malaria infection was demonstrated; the areas under the curves were 0.637 and 0.726, respectively.
There was no significant difference in the hemoglobin, leucocytes, lymphocytes, neutrophils, platelets counts, RDW, PDW, MPV and the parasite count (P=0.201) when the cases of P. falciparum and P. vivax were compared (Table 5). Plasmf = Blood film for P. falciparum.

Discussion
According to our present findings, PDW and MPV are the two most important haematological predictors of P. falciparum and P. vivax malaria infection. This is in line with a recent finding

Consent
The study was approved by the Institutional Review Board of the Medical College, University of Khartoum (3# 2015 1114).

Competing interests
No competing interests were disclosed.

Grant information
The author(s) declared that no grants were involved in supporting this work.
infection 8,15 . On the other hand, the significant differences observed in the haematological parameters between parasitemic Ugandan patients and non-parasitemic Ugandans were only observed in the monocyte and the platelet count 16  Nutritional deficiency and haemoglobinopathies were not investigated in the current study and have to be mentioned as study limitations. Haematological parameters formalaria-infested blood may vary depending on the level of malaria endemicity, presence of haemoglobinopathies and nutritional status 21,22 . Another limitation of the is that we relied on microscopy only for the malaria diagnosis. Some of negative controls may have had undetected parasitemia (submicroscopic parasitemia). We have previously observed 1.

If applicable, is the statistical analysis and its interpretation appropriate? Yes
Are all the source data underlying the results available to ensure full reproducibility? Yes

Are the conclusions drawn adequately supported by the results? Yes
No competing interests were disclosed.

Competing Interests:
Referee Expertise: Malaria and tropical medicine I have read this submission. I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard, however I have significant reservations, as outlined above. The objective of the study should be re-examined in the introduction. The terms "reliability" and "validity" should be differentiated. Reliability refers to the degree to which the measurement procedure can be reproduced.
Thus, if the objective is to study the validity of the hematological parameters to make a diagnosis, the methodology must be revised since these studies require a sample size that considers the prevalence and the expected sensitivity. Since this is a case-control study the objective should be consistent with this methodology.
It must be specified what a case is; What are the signs and symptoms of uncomplicated malaria?
It should be explained why 100 oil inmmersion fields were used as criteria to consider a thick blood film as negative. This does not correspond to international recommendations.
The calculation of the sample should be presented with the elements of a case-control study 5.

8.
The calculation of the sample should be presented with the elements of a case-control study i.e. case-control relation, expected O.R., frequency of outcomes in exposed and unexposed.
RESULTS. If the clinical history is available, the time of evolution of the disease could be compared, which may lead to differences in hematological profiles.
The results presented in the Tables 2 and 5 are confusing. It is not clear whether the averages or medians are compared and what was the statistical test of comparison. It is not explicitly stated which variables presented normal distribution.
The presumptive diagnosis of malaria with haematological analyzers does not seem to be the appropriate way of solving malaria diagnosis problems in endemic areas. In this sense, it is not clear whether the study intends to contribute to the malaria diagnosis using a method that would be non-specific.
The study contributes to show hematological differences between cases and controls and this should serve to identify early lesions, but clearly is no a useful method for malaria diagnosis.
There is no relevant discussion about the findings. The reference values of the hematological parameters according to age and sex are not presented; the comparisons must take into account this variation.
Some aspects related to platelet changes are analyzed but superficially. Leukocyte count and cell line analysis are neglected; no mention is made of the significance of alterations, monocytes and eosinophils are neglected. There are no correlations with parasitemias which is also a variable that can contribute significantly to the hematological alterations as a function of the greater or lesser inflammatory response.

If applicable, is the statistical analysis and its interpretation appropriate? Partly
Are all the source data underlying the results available to ensure full reproducibility? Partly Are the conclusions drawn adequately supported by the results? 1.

Are the conclusions drawn adequately supported by the results? Partly
No competing interests were disclosed. Mr Ali and colleagues present a study that revealed the mean platelet volume and platelet distribution width as good parameters to predict uncomplicated malaria parasite infection.The manuscript is well written and presented. The analysis of the results is adequate and correct. Some minor points should be reviewed: As the authors used several statistical tests to compare the results between cases and controls, it is important to make these tests explicit for each p-value presented in the tables, as a footnote.
The title of Table 5 is not clear. In fact, Table 5 compares means (SD), and non-median (IQR).

Is the work clearly and accurately presented and does it cite the current literature? Yes
Is the study design appropriate and is the work technically sound? Yes

If applicable, is the statistical analysis and its interpretation appropriate? Yes
Are all the source data underlying the results available to ensure full reproducibility? Yes

Are the conclusions drawn adequately supported by the results? Yes
No competing interests were disclosed.

Competing Interests:
I have read this submission. I believe that I have an appropriate level of expertise to confirm that it is of an acceptable scientific standard.