Evaluation of Sensitivity and Specificity of Malaria Detection by Hematology Analyzer using WBC separation and Non-WBC separation methods

Abstract

Recently, Depolarize-laser light in automated hematology analyzer has been found to be able to diagnose malaria by detecting the malaria pigment-containing White Blood Cell (WBC). This technique could be use for malaria detection during routine complete blood count analysis in the hospital which could save time and cost. However, in some studies found that a factor affecting the sensitivity and specificity of this technique was patient immunity to malaria. The immunity could induce WBC to better phagocytose malaria parasite. If patient had low immunity to malaria, occurring of phagocytosis might be lower. Thus, detection of malaria from pigment-containing WBC by this instrument would not be effective due to low sensitivity.

This research aimed at comparison of malaria detection by 3 fractions of a Hematology analyzer. WBC fraction and Leukocyte-rich plasma fraction compare, whole blood fraction, and a regular method. This study also suggests the appropriate and accuracy improvement of the instrument. Venous blood was collected from 223 patients attending malaria clinic for diagnosis malaria. Detection by hematology analyzer was conducted in the blood samples that WBC was separated or non-separated from whole blood. Thick blood film was used as a gold standard for the comparison.

The result showed that the specificity of the whole blood fraction was high (90%) but the

sensitivity was only 57%. This technique was thus not effectiveness for the diagnosis of malaria in

malaria suspected case. However, it could be helpful for malaria screening during the routine complete blood count due to its high specificity. Moreover, it could be a warning sign to laboratory technician, when an abnormal scatter plot (pigment containing WBC) was found. The microscopic examination of malaria was then confirmed.

Malaria parasite detection from the Leukocytes-rich plasma and WBC fraction had less sensitivity (41% and 47% respectively) when compared with whole blood fraction. Our assumption that concentration of WBC in the blood samples would increase the sensitivity of this instrument was rejected. Software modification of this instrument for malaria diagnosis in the future is necessary for example, the increase in the number of WBC injection during WBC counting process of more than 10,000 cells may help increase the sensitivity or the adaptation of the function to detect malaria directly from infected red blood cell.