Evaluating performance of multiplex real time PCR for the diagnosis of malaria at elimination targeted low transmission settings of Ethiopia

Abstract Background Malaria incidence has declined in Ethiopia in the past 10 years. Current malaria diagnostic tests, including light microscopy and rapid antigen-detecting diagnostic tests (RDTs) cannot reliably detect low-density infections. Studies have shown that nucleic acid amplification tests are highly sensitive and specific in detecting malaria infection. This study took place with the aim of evaluating the performance of multiplex real time PCR for the diagnosis of malaria using patient samples collected from health facilities located at malaria elimination targeted low transmission settings in Ethiopia. Methods A health facility-based, cross-sectional survey was conducted in selected malaria sentinel sites. Malaria-suspected febrile outpatients referred to laboratory for malaria testing between December 2019 and March 2020 was enrolled into this study. Sociodemographic information and capillary blood samples were collected from the study participants and tested at spot with RDTs. Additionally, five circles of dry blood spot (DBS) samples on Whatman filter paper and thick and thin smear were prepared for molecular testing and microscopic examination, respectively. Multiplex real time PCR assay was performed at Ethiopian Public Health Institute (EPHI) malaria laboratory. The performance of multiplex real time PCR assay, microscopy and RDT for the diagnosis of malaria was compared and evaluated against each other. Results Out of 271 blood samples, multiplex real time PCR identified 69 malaria cases as Plasmodium falciparum infection, 16 as Plasmodium vivax and 3 as mixed infections. Of the total samples, light microscopy detected 33 as P. falciparum, 18 as P. vivax, and RDT detected 43 as P. falciparum, 17 as P. vivax, and one mixed infection. Using light microscopy as reference test, the sensitivity and specificity of multiplex real time PCR were 100% (95% CI (93–100)) and 83.2% (95% CI (77.6–87.9)), respectively. Using multiplex real time PCR as a reference, light microscopy and RDT had sensitivity of 58% (95% CI 46.9–68.4) and 67% (95% CI 56.2–76.7); and 100% (95% CI 98–100) and 98.9% (95% CI 96–99.9), respectively. Substantial level of agreement was reported between microscopy and multiplex real time PCR results with kappa value of 0.65. Conclusions Multiplex real-time PCR had an advanced performance in parasite detection and species identification on febrile patients’ samples than did microscopy and RDT in low malaria transmission settings. It is highly sensitive malaria diagnostic method that can be used in malaria elimination programme, particularly for community based epidemiological samples. Although microscopy and RDT had reduced performance when compared to multiplex real time PCR, still had an acceptable performance in diagnosis of malaria cases on patient samples at clinical facilities.