Sensitivity and specificity of HAT Sero-K-SeT, a rapid diagnostic test for serodiagnosis of sleeping sickness caused by Trypanosoma brucei gambiense: a case-control study

Abstract

Background: Human African trypanosomiasis (HAT) is a life-threatening infection affecting rural populations in sub-Saharan Africa. Large-scale population screening by antibody detection with the Card Agglutination Test for Trypanosomiasis (CATT)/Trypanosoma brucei (T b) gambiense helped reduce the number of reported cases of gambiense HAT to fewer than 10 000 in 2011. Because low case numbers lead to decreased cost-effectiveness of such active screening, we aimed to assess diagnostic accuracy of a rapid serodiagnostic test (HAT Sero-K-SeT) applicable in primary health-care centres. Methods: In our case-control study, we assessed participants older than 11 years who presented for HAT Sero-K-SeT and CATT/T b gambiense at primary care centres or to mobile teams (and existing patients with confirmed disease status at these centres) in Bandundu Province, DR Congo. We defined cases as patients with trypanosomes that had been identified in lymph node aspirate, blood, or cerebrospinal fluid. During screening, we recruited controls without previous history of HAT or detectable trypanosomes in blood or lymph who resided in the same area as the cases. We assessed diagnostic accuracy of three antibody detection tests for gambiense HAT: HAT Sero-K-SeT and CATT/T b gambiense (done with venous blood at the primary care centres) and immune trypanolysis (done with plasma at the Institute of Tropical Medicine, Antwerp, Belgium). Findings: Between June 6, 2012, and Feb 25, 2013, we included 134 cases and 356 controls. HAT Sero-K-SeT had a sensitivity of 0·985 (132 true positives, 95% CI 0·947–0·996) and a specificity of 0·986 (351 true negatives, 0·968–0·994), which did not differ significantly from CATT/T b gambiense (sensitivity 95% CI 0·955, 95% CI 0·906–0·979 [128 true positives] and specificity 0·972, 0·949–0·985 [346 true negatives]) or immune trypanolysis (sensitivity 0·985, 0·947–0·996 [132 true positives] and specificity 0·980, 0·960–0·990 [349 true negatives]). Interpretation: The diagnostic accuracy of HAT Sero-K-SeT is adequate for T b gambiense antibody detection in local health centres and could be used for active screening whenever a cold chain and electricity supply are unavailable and CATT/T b gambiense cannot be done. Funding: European Commission FP7 (NIDIAG, Grant Agreement 260260) and WHO Control of Neglected Tropical Diseases, Innovative and Intensified Disease Management.