Performance evaluation of the automated haematology analyzer XN-31 for malaria diagnosis in Plasmodium vivax-dominant regions of Thailand

Abstract

Background: The innovative automated haematology analyzer XN-31 has demonstrated comparable or, in some settings, superior performance to microscopy and rapid diagnostic tests (RDTs), and good concordance with polymerase chain reaction (PCR) in various endemic areas, particularly where Plasmodium falciparum is prevalent. The XN-31 applies the principle of flow cytometry to measure cell size and nucleic acid content to generate a two-dimensional cytogram in approximately one minute. This technique identifies P. falciparum and other species and provides % parasitaemia (MI-RBC%). This study evaluated the diagnostic performance of the XN-31 in Thailand, where Plasmodium vivax is predominant. Methods: From November 2019 to June 2022, 349 patients suspected of having malaria were enrolled at the Hospital for Tropical Medicine, Mahidol University, Bangkok. Blood samples were collected via venipuncture and analysed using the XN-31, thick and thin film microscopy, RDTs, and PCR. The qualitative diagnostic accuracy of the XN-31 for detecting malaria parasites and identifying their species was compared with other methods. The quantitative diagnostic ability of the XN-31 was assessed by correlating the MI-RBC% values with % parasitaemia from thin film microscopy. Results: Among the 349 samples, 125 were positive according to thick film microscopy (103 P. vivax, 17 P. falciparum, 3 Plasmodium knowlesi, 1 Plasmodium ovale, and 1 Plasmodium malariae). The XN-31 demonstrated 98.4% sensitivity relative to thick film microscopy as the reference method, outperforming the RDT (88.0%). All XN-31-positive samples were confirmed via PCR. The information on the parasite species, flagged as ‘Malaria? (P.f)’ or ‘Malaria? (others)’ by the XN-31, which depicts P. falciparum or other Plasmodium species, respectively, matched 100% with microscopic determination. The quantitative performance of the XN-31 was strongly correlated with that of thin film microscopy (correlation coefficient [r] = 0.903). Conclusions: This is the first study to confirm the accuracy of the XN-31 for both qualitative and quantitative diagnoses in a malaria-endemic region dominated by P. vivax. The XN-31 is expected to be a useful diagnostic tool in similar endemic regions.