ABSTRACT
BACKGROUND: Despite some problems related to accuracy and applicability, malaria rapid diagnostic tests (RDTs), are currently considered the best option in areas with limited laboratory services for improving case management and reducing over-treatment. However, their performance must be established taking into the account the particularities of each endemic area. In the Democratic Republic of Congo, the validity of Optimal-IT(®) and Paracheck-Pf(®), respectively based on the detection of lactate dehydrogenase and histidine-rich protein-2, was assessed at primary health care level (PHC). METHODS: This was a two-stage cluster randomized survey, conducted in one health centre in 12 health zones in Kinshasa city. All patients with malaria presumptive diagnosis were eligible. Gold standard was microscopy performed by experts from the parasitology unit, Kinshasa University. RESULTS: 624 patients were enrolled. 53.4% (95% CI: 49.4-57.3) owed a bed net, obtained in 74.5% of cases (95% CI: 69.4-79.1) through community-based distribution by the National Malaria Control Programme. Microscopy expert reading confirmed 123 malaria cases (19.7%; 95% CI: 16.7-23.1). Overall sensitivity were 79.7% (95% CI: 72.4-86.8), 87.8% (95% CI: 81.9-93.6) and 86.2% (95% CI: 79.9-92.3), respectively, for Optimal-IT(®), Paracheck-Pf(®) and microscopy performed at PHC. Specificity was 97.0% (95% CI: 95.5-98.5), 91.6% (95% CI: 89.1-94.0) and 49.1% (95% CI: 44.7-53.4). The proportion of confirmed cases seemed similar in under-fives compared to others. Any treatment prior to the current visit was a predictor for malaria (AOR: 2.3; 95% CI: 1.5-3.5), but not malaria treatment (AOR: 0.87; 95% CI: 0.4-1.8). Bed net ownership tended to protect against malaria (AOR: 0.67; 95% CI: 0.45-0.99). CONCLUSION: Although microscopy is considered as the "gold standard" for malaria diagnosis at point of care level, this study showed that its accuracy may not always be satisfactory when performed in health centres.
