Background: Rapid diagnostic test (RDT) kits have been useful tools to screen for the presence of infection with malaria parasites. Despite the improvement, false results from RDTs present a greater challenge. The need for quality test kits is desirable. We evaluated the diagnostic accuracy of three malaria RDTs.
Method: The team consented and enrolled 754 participants from the two major public hospitals in Kintampo districts of Ghana from June 2014 to August 2014. Venous blood samples were obtained by trained personnel and samples were screened for malaria using CareStart and SD Bioline HRP2 and HRP2 with pLDH based RDTs with blood slides for malaria microscopy as "gold standard". Geometric mean parasite densities were estimated and parasite densities were used to estimate the quantitative limits of the RDTs. The sensitivities, specificities and other performance criteria were calculated using statistical analytical software.
Result: The median age of participants was 21 (range 5-31) years. There were 28.6% (215/752) were males and 71.4% (537/752) were females. Comparing with microscopy, the sensitivity, specificity, positive predictive value, negative predictive value and the ROC were for CareStart (HRP2), 98.2%, 66.5%, 82.6%, 95.6%, 0.82; for CareStart (HRP2/pLDH) 98.2%, 66.5%, 82.6%, 95.6%, 0.82 and for SD-Bioline (HRP2/pLDH) RDTs 98.2%, 69.2%, 84.2%, 96.0%, 0.84 respectively. The performance for all the kits were acceptable at a cut-off of 25 or more parasites/μl of blood.
Conclusions: The diagnostic performance of the three malaria RDTs was acceptable, according to the World Health Organisation criteria, to detect densities ≥25 parasite/μl of blood. The RDTs with HRP2/pLDH targets were comparable to those with only HRP2 and could successfully substitute current and commonly used HRP2-based RDTs.