Comparison of methodologies for microscopic malaria diagnosis

Introduction: As part of the pre-elimination plan for malaria in Colombia, it has been proposed to develop activities within the line of work: “Improve access and quality of malaria diagnosis”. Objective: To compare the methodology recommended by PAHO/WHO with that used in Colombia for the diagnosis of malaria. Materials and methods: Samples were collected and 88 slides were prepared for malaria diagnosis, under different scenarios according to the parameters to be evaluated. After duplicate mycroscopic reading, the respective variance calculations were performed for all possible staining comparisons with the two methods used (thick smear, combined thick smear), according to the staining (modified Romanowsky or Giemsa), with the result variable being the parasite density (500, 1,000, 5,000 and 10,000 parasites/μl of blood). Results: A Cohen kappa index of inter-rater agreement of 0.923 (95% CI: 0.768-1.078) was obtained. None of the factors (A: stain, B: methodology) or interactions (AB) had a statistically significant effect on the results with a 95% confidence level. Conclusion: Based on the results of the study, the preparation of two thick smears in the same slide stained with the modified Romanowsky stain is a suitable methodology for the diagnosis of malaria in Colombia, due to its technical characteristics, of storage, low cost, use and care.