METHODS: The protocol of this systematic review was registered in the PROSPERO International Prospective Register of Systematic Reviews (ID = CRD42020204770). Studies reporting the misidentification of P. knowlesi as P. malariae by microscopy and confirmation of this by molecular methods in MEDLINE, Web of Science and Scopus were reviewed. The risk of bias in the included studies was assessed using the Quality Assessment of Diagnostic Accuracy Studies (QUADAS). The pooled prevalence and 95% confidence interval (CI) of the misidentification of P. knowlesi as P. malariae by microscopy were estimated using a random effects model. Subgroup analysis of the study sites was performed to demonstrate any differences in the misidentification rates in different areas. Heterogeneity across the included studies was assessed and quantified using Cochran's Q and I2 statistics, respectively. Publication bias in the included studies was assessed using the funnel plot, Egger's test and contour-enhanced funnel plot.
RESULTS: Among 375 reviewed studies, 11 studies with a total of 1569 confirmed P. knowlesi cases in humans were included. Overall, the pooled prevalence of the misidentification of P. knowlesi as P. malariae by microscopy was estimated at 57% (95% CI 37-77%, I2: 99.3%). Subgroup analysis demonstrated the highest rate of misidentification in Sawarak, Malaysia (87%, 95% CI 83-90%, I2: 95%), followed by Sabah, Malaysia (85%, 95% CI 79-92%, I2: 85.1%), Indonesia (16%, 95% CI 6-38%), and then Thailand (4%, 95% CI 2-9%, I2: 95%).
CONCLUSION: Although the World Health Organization (WHO) recommends that all P. malariae-positive diagnoses made by microscopy in P. knowlesi endemic areas be reported as P. malariae/P. knowlesi malaria, the possibility of microscopists misidentifying P. knowlesi as P. malariae is a diagnostic challenge. The use of molecular techniques in cases with malariae-like Plasmodium with high parasite density as determined by microscopy could help identify human P. knowlesi cases in non-endemic countries.