Accuracy of urine circulating cathodic antigen (CCA) test for Schistosoma mansoni diagnosis in different settings of Côte d'Ivoire

Abstract

Background: Promising results have been reported for a urine circulating cathodic antigen (CCA) test for the diagnosis of Schistosoma mansoni. We assessed the accuracy of a commercially available CCA cassette test (designated CCA-A) and an experimental formulation (CCA-B) for S. mansoni diagnosis.

Methodology: We conducted a cross-sectional survey in three settings of Côte d'Ivoire: settings A and B are endemic for S. mansoni, whereas S. haematobium co-exists in setting C. Overall, 446 children, aged 8-12 years, submitted multiple stool and urine samples. For S. mansoni diagnosis, stool samples were examined with triplicate Kato-Katz, whereas urine samples were tested with CCA-A. The first stool and urine samples were additionally subjected to an ether-concentration technique and CCA-B, respectively. Urine samples were examined for S. haematobium using a filtration method, and for microhematuria using Hemastix dipsticks.

Principal findings: Considering nine Kato-Katz as diagnostic 'gold' standard, the prevalence of S. mansoni in setting A, B and C was 32.9%, 53.1% and 91.8%, respectively. The sensitivity of triplicate Kato-Katz from the first stool and a single CCA-A test was 47.9% and 56.3% (setting A), 73.9% and 69.6% (setting B), and 94.2% and 89.6% (setting C). The respective sensitivity of a single CCA-B was 10.4%, 29.9% and 75.0%. The ether-concentration technique showed a low sensitivity for S. mansoni diagnosis (8.3-41.0%). The specificity of CCA-A was moderate (76.9-84.2%); CCA-B was high (96.7-100%). The likelihood of a CCA-A color reaction increased with higher S. mansoni fecal egg counts (odds ratio: 1.07, p<0.001). A concurrent S. haematobium infection or the presence of microhematuria did not influence the CCA-A test results for S. mansoni diagnosis.

Conclusion/significance: CCA-A showed similar sensitivity than triplicate Kato-Katz for S. mansoni diagnosis with no cross-reactivity to S. haematobium and microhematuria. The low sensitivity of CCA-B in our study area precludes its use for S. mansoni diagnosis.

Figure 1. STARD-flowchart showing study participation, stratified by epidemiological setting.

Flowchart detailing the study participation and adherence of children for multiple stool and urine submissions for diagnosis of S. mansoni and S. haematobium infections in Azaguié, south Côte d'Ivoire, in October and November 2010. According to nine Kato-Katz thick smear examinations, the prevalence of S. mansoni in setting A, B and C was 32.9%, 53.1% and 91.8%, respectively. Of note, S. haematobium is co-endemic in setting C.

Figure 2. Correlation between Kato-Katz and CCA for S. mansoni diagnosis.

Figure showing the correlation between the prevalence and intensity (stratified by intensity class) of S. mansoni infections, as determined by a single or triplicate CCA-A (light blue and dark blue bar, respectively), and a single CCA-B (yellow bar), stratified by study setting. According to nine Kato-Katz thick smear examinations, the prevalence of S. mansoni in setting A, B and C was 32.9%, 53.1% and 91.8%, respectively. In setting C, S. haematobium is co-endemic.