EVALUATION OF 6 COMMERCIAL KITS FOR DETECTION OF HUMAN-IMMUNOGLOBULIN-M ANTIBODIES TO TOXOPLASMA-GONDII

As a result of reports received by the Food and Drug Administration (F DA) of false-positive results obtained with FDA-cleared in vitro diagn ostic kits for the detection of Toxoplasma-specific human immunoglobul in M (IgM) antibodies, an FDA-sponsored evaluation of six kits was per formed, A battery of 258 serum specimens, including 30 specimens drawn 1 to 5 months after initial Toxoplasma infection and 228 specimens fr om Toxoplasma IgG-positive individuals, Toxoplasma IgG-negative indivi duals, rheumatoid factor-positive persons, and persons determined to b e Toxoplasma IgM positive by commercially available assays, was assemb led, randomly assorted, and coded. The battery was tested at the FDA w ith six commercially available hits, at the Pale Alto Medical Foundati on (PAMF) by the PAMF double-sandwich IgM enzyme-linked immunosorbent assay (PAMF IgM ELISA), and at the Centers for Disease Control and Pre vention (CDC) by the CDC EIA IgM. The results of the PAMF IgM ELISA th at were obtained with the battery were considered to be the ''gold sta ndard'' for this study; specificity rates were computed by considering the PAMF results to be 100% specific, Sensitivity and specificity rat es were found to be as follows: CDC EIA IgM, 100 and 99.1%, respective ly; Abbott IMx Toro IgM, version 1, 100 and 77.5%, respectively; Abbot t IMx Toro IgM, version 2, 93.3 and 97.3%, respectively; Abbott Toxo-M EIA, 100 and 84.2%, respectively; BioMerieux Vitek VIDAS Toro IgM, 10 0 and 98.6%, respectively; BioWhittaker Toxocap-M, 100 and 95.9%, resp ectively; Gull Toxo IgM, 97 and 85.6%, respectively; and Sanofi Diagno stics Pasteur Platelia Toro IgM, 100 and 96.8%, respectively, Although the extent of false-positive reactions with these kits cannot be calc ulated because the study was retrospective and sample choices were bia sed, the results may be useful as an indicator of the relative specifi cities of these kits.