The objectives of this study were to conduct a further evaluation of performance characteristics (sensitivity/specificity, predictive values, cross-reactivity) of PanBio
Dengue IgM (
IgM-EIA test), particularly during non-epidemic periods in New Caledonia, and (ii) to evaluate an alternative test, Pentax
Dengue IgM-Particle Agglutination (PA-lgM) test. A total of 1,808 samples were first tested with the
IgM-EIA test and reactive specimens were then re-tested with
IgM-PA test. Sensitivity and specificity were measured on a prospective mode from 2005 and 2006. Other etiologies were also investigated to confirm the non-specific reactive results. One hundred fifty-three samples were initially reactive with
IgM-EIA test. Of these, 147 were classified as non specific and only 16 were reactive with the particle agglutination test (89.1% reduction of this interference). The specificity and positive predictive value of the ELISA test was 91.8% and 5.8%, respectively. The extrapolated specificity and positive predictive value for the particle agglutination test was 99.1% and 33.3%, respectively.
Hepatitis A was identified as a major source of false positive, followed by
rheumatoid factor and
leptospirosis. Sensitivity of both tests was100% on samples taken from the fifth day of the disease.