Malaria Rapid Diagnostic Tests and Malaria Microscopy for Guiding Malaria Treatment of Uncomplicated Fevers in Nigeria and Prereferral Cases in 3 African Countries

BACKGROUND  The World Health Organization recommends that malaria treatment be based on demonstration of the infecting Plasmodium parasite specie. Malaria rapid diagnostic tests (RDTs) are recommended at community points of care because they are accurate and rapid. We report on parasitological results in a malaria study in selected rural communities in 3 African countries. METHODS  In Nigeria, community health workers (CHWs) performed RDTs (SD-Bioline) and thick blood smears on all children suspected to have malaria. Malaria RDT-positive children able to swallow received artemisinin-based combination therapy (Coartem). In all countries, children unable to take oral drugs received prereferral rectal artesunate irrespective of RDT result and were referred to the nearest health facility. Thick blood smears and RDTs were usually taken at hospital admission. In Nigeria and Burkina Faso, RDT cassettes and blood smears were re-read by an experienced investigator at study end. RESULTS  Trained CHWs enrolled 2148 children in Nigeria. Complete parasitological data of 1860 (86.6%) enrollees were analyzed. The mean age of enrollees was 30.4 ± 15.7 months. The prevalence of malaria parasitemia in the study population was 77.8% (1447/1860), 77.6% (1439/1855), and 54.1% (862/1593) by RDT performed by CHWs vs an expert clinical research assistant vs microscopy (gold standard), respectively. Geometric mean parasite density was 6946/µL (range, 40-436 450/µL). There were 49 cases of RDT false-negative results with a parasite density range of 40-54 059/µL. False-negative RDT results with high parasitemia could be due to non-falciparum infection or result from a prozone effect. Sensitivity and specificity of SD-Bioline RDT results as read by CHWs were 94.3% and 41.6%, respectively, while the negative and positive predictive values were 86.1% and 65.6%, respectively. The level of agreement in RDT reading by the CHWs and experienced research staff was 86.04% and κ statistic of 0.60. The malaria parasite positivity rate by RDT and microscopy among children with danger signs in the 3 countries was 67.9% and 41.8%, respectively. CONCLUSIONS  RDTs are useful in guiding malaria management and were successfully used for diagnosis by trained CHWs. However, false-negative RDT results were identified and can undermine confidence in results and control efforts.