Accuracy of Individual Rapid Tests for Serodiagnosis of Gambiense Sleeping Sickness in West Africa

Abstract

Individual rapid tests for serodiagnosis (RDT) of human African trypanosomiasis (HAT) are particularly suited for passive screening and surveillance. However, so far, no large scale evaluation of RDTs has been performed for diagnosis of Trypanosoma brucei gambiense HAT in West Africa. The objective of this study was to assess the diagnostic accuracy of 2 commercial HAT-RDTs on stored plasma samples from West Africa. SD Bioline HAT and HAT Sero-K-Set were performed on 722 plasma samples originating from Guinea and Côte d’Ivoire, including 231 parasitologically confirmed HAT patients, 257 healthy controls, and 234 unconfirmed individuals whose blood tested antibody positive in the card agglutination test but negative by parasitological tests. Immune trypanolysis was performed as a reference test for trypanosome specific antibody presence. Sensitivities in HAT patients were respectively 99.6% for SD Bioline HAT, and 99.1% for HAT Sero-K-Set, specificities in healthy controls were respectively 87.9% and 88.3%. Considering combined positivity in both RDTs, increased the specificity significantly (p≤0.0003) to 93.4%, while 98.7% sensitivity was maintained. Specificities in controls were 98.7–99.6% for the combination of one or two RDTs with trypanolysis, maintaining a sensitivity of at least 98.1%. The observed specificity of the single RDTs was relatively low. Serial application of SD Bioline HAT and HAT Sero-K-Set might offer superior specificity compared to a single RDT, maintaining high sensitivity. The combination of one or two RDTs with trypanolysis seems promising for HAT surveillance. Screening for gambiense human African trypanosomiasis (HAT) or sleeping sickness is traditionally based on detection of trypanosome specific antibodies in blood. Whereas the card agglutination test is particularly suited for mass screening, individual rapid serodiagnostic tests (RDTs) are rather adapted for use in peripheral health-care centres. Two RDTs have been commercialized recently, and we assessed their diagnostic accuracy on stored plasma samples from West Africa. Immune trypanolysis was performed as a laboratory reference test for antibody presence. Although sensitivity for serodiagnosis of HAT in West Africa was high for both RDTs, their specificity was only 88%. Taking into account the high number of false positive test results, combined seropositivity in both RDTs was considered, raising specificity to 93%. Serial application of two RDTs should therefore be considered as an option for passive case finding, especially in settings with low HAT prevalence. A combination of one or two RDTs with immune trypanolysis further improved specificity for HAT to 99%, while maintaining sensitivity at 99% and seems promising for HAT surveillance.