Comparison of microscopy and PCR for the detection of human Plasmodium species and Plasmodium knowlesi in southern Myanmar

To determine the distribution of Plasmodium (P) species including Plasmodium knowlesi and to compare the specificity and sensitivity of microscopy with nested PCR in malaria diagnosis. The study was conducted in Kawthaung, southern Myanmar. Ninety clinically suspected malaria patients were screened for malaria by Giemsa stained microscopy and confirmed by nested PCR. Among the participants, 57 (63.3%) were positive and 33 (36.7%) were negative by microscopy. Of positive samples, 39 (68.4%) were Plasmodium falciparum, 17 (29.8%) Plasmodium vivax and 1 (1.8%) Plasmodium malariae, whereas 59-amplified by PCR were 40 (67.8%), 18 (30.5%) and 1 (1.7%) respectively. PCR amplified 2 microscopy negative samples. Two samples of P. falciparum detected by microscopy were amplified as P. vivax and vice versa. All samples were negative for Plasmodium ovale, P. knowlesi and mixed infections. Microscopy had a very good measure of agreement (κ = 0.95) compared to nested PCR. Sensitivity and specificity of microscopy for diagnosis of P. falciparum were 92.5% (95% CI: 79.6–98.4) and 96.0% (95% CI: 86.3–99.5) respectively, whereas for P. vivax were 83.3% (95% CI: 58.6–96.4) and 97.2% (95% CI: 90.3–99.7). P. knowlesi was not detected by both microscopy and PCR. Giemsa stained microscopy can still be applied as primary method for malaria diagnosis and is considered as gold standard. As to the lower sensitivity of microscopy for vivax malaria, those with previous history of malaria and relapse cases should be diagnosed by RDT or PCR combined with microscopy. Inaccuracy of species diagnosis highlighted the requirement of training and refresher courses for microscopists.