Thrombocytopenia as a Diagnostic Marker for Malaria in Patients with Acute Febrile Illness

Background Thrombocytopenia is the most common hematological abnormality in patients with acute malaria. This study aimed to determine the role of thrombocytopenia as a diagnostic marker for malaria in patients with acute febrile illness. Method A cross-sectional health facility-based study was conducted on 423 consecutively selected acute febrile patients at Ataye District Hospital from February to May 2019 GC. A complete blood count and malaria microscopy were performed for each acute febrile patient. ROC curve analysis was performed to calculate sensitivity, specificity, positive predictive value, and negative predictive value of platelet count in predicting malaria. A P ≤ 0.05 was considered statistically significant. Result Out of the 423 acute febrile patients, 73 (17.3%) were microscopically confirmed malaria cases and the rest 350 (82.7%) patients had negative blood film results. Of the microscopically confirmed malaria cases, 55 (75.34%) were P. vivax and 18 (24.66%) were P. falciparum. The prevalence of thrombocytopenia among malaria patients (79.5%) was significantly higher than those in malaria negative acute febrile patients (13.7%), P < 0.001. About 67% malaria-infected patients had mild to moderate thrombocytopenia and 12.3% had severe thrombocytopenia. The ROC analysis demonstrated platelet counts <150,000/μl as an optimal cutoff value with 0.893 area under the curve, 79.5% sensitivity, 86.3% specificity, 95.3% negative predictive value, and 54.7% positive predictive value to predict malaria. Conclusion Malaria is still among the major public health problems in the country. Thrombocytopenia is a very good discriminatory test for the presence or absence of malaria with 79.5% sensitivity and 86.3% specificity. Therefore, this may be used in addition to the clinical and microscopic parameters to heighten the suspicion of malaria.


Background
Malaria is a life-threatening infectious disease caused by protozoan parasites of the Plasmodium species. Globally, there were an estimated 228 million malaria cases and 405, 000 deaths in 2018, and 93% of the malaria cases and 94% of the deaths were in the African Region [1,2]. e sub-Saharan Africa region was the most affected area contributing the largest burden of malaria morbidity and mortality. In Ethiopia, malaria is a serious public health problem with around 68% of the population is at risk of malaria. Approximately, 2.9 million malaria cases and 4,782 deaths were reported in 2016. e most dominant plasmodium species are Plasmodium falciparum and Plasmodium vivax [3][4][5].
ose hematological alterations may vary with level of malaria endemicity, background hemoglobinopathy, nutritional status, demographic factors, and malaria immunity [10]. rombocytopenia is the most common hematological abnormality in patients with acute malaria [9,11]. Severe thrombocytopenia is associated with increased risk of mortality in both child and adult patients with P. falciparum and P. vivax infections [12]. e mechanism of thrombocytopenia in malaria is not clear, but the speculated mechanisms are an increase in the consumption or destruction of platelets, or suppression of thrombopoiesis, or a combination of both. Increased consumption or destruction of platelets during malarial infection occurs due to antibodymediated platelet destruction, disseminated intravascular coagulation (DIC), pooling within the reticuloendothelial system, sequestration in the microcirculation, oxidative stress, and malaria-mediated apoptosis [13][14][15][16].
ere are some studies that reported the presence of thrombocytopenia is a predictor of acute malaria in patients with acute febrile illness in endemic areas [17][18][19][20]. However, there is scarce information on usefulness of thrombocytopenia in predicting acute malaria in Ethiopia. erefore, this study is aimed to determine the role of thrombocytopenia as a diagnostic marker for malaria in patients with acute febrile illness in Ethiopia.

Study Design, Area, and Population.
A cross-sectional health facility-based study was conducted at Ataye District Hospital from February to May 2019 GC. e Hospital is found in Ataye Town, Ataye District, North Shewa Zone of the Amhara Region, Ethiopia. Ataye Town is located at an elevation of 1468 meters above sea level and about 270 km away from the capital city Addis Ababa. Ataye District is one of the hot and malaria endemic areas in Amhara Region, Ethiopia. Malaria is one of the major health problems of the district. Among the plasmodium species, P. vivax and P. falciparum are main plasmodium species responsible for malaria infection in the district.
A total of 423 consecutively selected acute febrile patients were enrolled from Ataye District Hospital. Acute febrile patients with pregnancy, HIV, known hematological malignancies, bleeding disorders, and antimalarial therapy were excluded from the study.

Data Collection
Procedure. Semistructured questionnaires were used to collect sociodemographic and clinical data of the study participants by trained clinical nurses. A complete blood count (CBC) and malaria microscopy were performed for each acute febrile patient. Sysmex KX-21N (Sysmex Corporation, Kobe, Japan) automated hematological analyzer was used to determine the complete blood count (total white blood cell count, red blood cell, and platelet count). Giemsa-stained thin and thick blood smears were used to diagnose malaria. in smears were considered positive for malaria if one or more malarial parasites were seen and negative if no asexual form of Plasmodium was detected in a minimum of 200 oil immersion fields. On thick blood smears, detection of any levels of asexual forms of malarial parasites was considered malaria-positive and if no parasites were seen after examining 1000 white blood cells labeled as negative. e blood smear examination and complete blood count were performed by two experienced medical laboratory scientists. ese laboratory professionals had trainings on malaria microscopy. Standard operating procedures (SOPs) were strictly followed in each step to maintain the quality of the laboratory results.

Data Analysis.
e data were entered and analyzed using a Statistical Package for the Social Science (SPSS) Version 20 statistical software. We used the t-test for continuous variables and the chi-square test for categorical variables to compare means and proportions. ROC curve analysis was performed to calculate sensitivity, specificity, positive predictive value, and negative predictive value of platelet count in predicting malaria. In all cases, a P value less than 0.05 was considered statistically significant.

Ethical Considerations.
Ethical approval was obtained from the Research and Ethics Review Committee of College of Medicine and Health Sciences, Wollo University. Permission to conduct the study was also obtained from Ataye District Hospital. Written informed consent from adult participants and assent for children under 18 years of age were obtained before enrolment into the study. To ensure confidentiality, participants' data were linked to a code number. Any abnormal test results of participants were communicated to the concerned body in the hospital.

Sociodemographic and Clinical Characteristics of Participants.
A total of 423 subjects with acute febrile illness were enrolled in this study.

Malaria and rombocytopenia.
Out of the 423 acute febrile patients, 73 (17.3%) were microscopically confirmed malaria cases and 350 (82.7%) had negative blood films (Table 1). Of the microscopically confirmed malaria cases, 55 (75.34%) were P. vivax positive and 18 (24.66%) were P. falciparum. ere was no statistically significant difference in malaria infection between male and female acute febrile patients (14.7% and 19.4%, respectively). Malaria infection was more prevalent in rural residents (22.8%) than urban residents (11.1%), (P � 0.002). e mean platelet count was significantly lower in patients with malaria infection than patients without malaria infection (P < 0.001). e prevalence of thrombocytopenia among malaria patients (79.5%) was significantly higher than those of malaria negative acute febrile patients (13.7%), P < 0.001. Regarding to the severity of thrombocytopenia, about 67% malaria-infected patients had mild to moderate thrombocytopenia and 12.3% had severe thrombocytopenia ( Table 2). e prevalence of severe thrombocytopenia was higher in P. falciparum (38.9%) than in P. vivax (3.6%) infected study participants. Most of the malaria negative acute febrile patients (86.3%) had normal platelet count (Figure 1).

Diagnostic Values of Platelet Count.
In the ROC analysis, the area under the curve (AUC) of the platelet count was 0.893 (95% CI: 0.847-0.938; P < 0.001), the optimum cutoff point of platelet count to differentiate malaria infection from acute febrile patients was less than 150 × 10 3 /µl with sensitivity of 79.5% and specificity of 86.3% (Figure 2 and Table 3).

Discussion
Malaria is one of the high burden diseases in developing countries. Ethiopia is one of the sub-Saharan countries highly endemic to malaria where an estimated 68% of the population lives in malarious areas [4,5]. In the present study, the prevalence of microscopically confirmed malaria cases was 17.3% which is similar to studies conducted in the Zeway Health Center, Ethiopia (17%) [22], Gurage Zone, Southern Ethiopia (18.3%) [23], East Nile locality of Khartoum State (18.5%) [24], and Kenya (15.5%) [25]. However, the finding of this study is lower than studies conducted in Arba Minch, Southern Ethiopia (27.6%) [20], Kersa Woreda, Ethiopia (43.8%) [26], Zaria, Nigeria (45.4%) [27], and New Delhi (24%) [18] and higher than studies conducted in North Shoa, Ethiopia (8.4%), and District Dir Lower, Pakistan (12.2%) [28,29]. e observed variation might be due to seasonal climatic condition and altitude difference that might influence breeding of malaria vector and community awareness about malaria transmission and control. Among the confirmed malaria cases, the predominant Plasmodium species was P. vivax (75%), followed by P.falciparum (25%).
is was in agreement with other studies [23,29,30], but other studies reported that the most prevalent species was P. falciparum [26,28].
In this study, mean platelet counts were significantly reduced in malaria-infected patients than those with nonmalaria. rombocytopenia occurred in 79.5% of malariainfected patients. ese findings imply that thrombocytopenia may be a marker of Plasmodium infection. e association of thrombocytopenia and malaria infection was in agreement with previous studies [9,20,[31][32][33]. No statistically significant difference in the prevalence of thrombocytopenia was observed between P. falciparum (83.3%) and P. vivax (78.2%) infected patients (P � 0.7482) which is similar to a study done by Kassa et al. [34]. In contrast to our study, Patel et al. showed significantly higher incidence of thrombocytopenia in P. falciparum than P. vivax [11] and Shaikh et al. showed significantly higher incidence of thrombocytopenia in P. vivax infected patients [33]. rombocytopenia in malaria infection might occur due to an increase in the consumption or destruction of platelets, or suppression of thrombopoiesis, or a combination of both. e suggested mechanism of accelerated clearance or consumption of platelets during malarial infection includes disseminated intravascular coagulation (DIC), immunemediated destruction, pooling within the reticuloendothelial system, sequestration in the microcirculation, and malariamediated apoptosis [13][14][15][16]. In contrast to this study, a study was reported no association between the mean platelet count and irritable bowel syndrome [35]. A study was reported no difference between mean platelet count and TSH levels between the malignant thyroid nodule group and benign nodule group [36]. ere are a lot of possible reasons for the increase or decrease of the peripheral platelet count. Among the different variables that can determine the mean peripheral platelet count, platelet production rate, mean platelet survival/life span, and the size of the exchangeable splenic platelet pool can be mentioned [37]. In contrast to the relationship between malaria and thrombocytopenia in the present study, a study reported by Sincer et al. showed that there was an inverse relationship between the number of blood eosinophil count and the severity of acute coronary syndrome (ACS) in elderly patients [38].
In this study, the ROC analysis demonstrated platelet counts <150,000/μl as an optimal cutoff value with 79.5%

Conclusion
Malaria is still among the major public health problems in the country. e prevalence of thrombocytopenia was significantly higher among malaria patients than malaria negative acute febrile patients in Ataye Hospital. rombocytopenia is a good discriminatory test for the presence or absence of malaria with 79.5% sensitivity and 86.3% specificity. erefore, this may be used in addition to the clinical and microscopic parameters to heighten the suspicion of malaria.
Data Availability e authors confirm that all data underlying the findings are fully available without restriction. All relevant data are within the manuscript.

Ethical Approval
Ethical approval was obtained from the Research and Ethics Review Committee of College of Medicine and Health Sciences, Wollo University. Permission to conduct the study was also obtained from Ataye District Hospital.

Conflicts of Interest
e authors declare that they have no conflicts of interest.

Authors' Contributions
AG and YE were involved in the conception, design, analysis, interpretation, report writing, and manuscript writing. AG, DGF, GH, and TF had been involved in the design, analysis, and critically reviewing the manuscript. All the authors read and approved the final manuscript. Journal of Tropical Medicine 5