Résumé
ABSTRACT The molecular detection of Toxoplasma gondii DNA is a key tool for the diagnosis of disseminated and congenital toxoplasmosis. This multicentric study from the Molecular Biology Pole of the French National Reference Center for toxoplasmosis aimed to evaluate Toxoplasma gondii Real-TM PCR kit (Sacace). The study compared the analytical and clinical performances of this PCR assay with the reference PCRs used in proficient laboratories. PCR efficiencies varied from 90% to 112%; linearity zone extended over four log units (R 2 > 0.99) and limit of detection varied from 0.01 to ≤1 Tg/mL depending on the center. Determined on 173 cryopreserved DNAs from a large range of clinical specimens, clinical sensitivity was 100% [106/106; 95 confidence interval (CI): 96.5%–100%] and specificity was 100% (67/67; 95 CI: 94.6%–100%). The study revealed two potential limitations of the Sacace PCR assay: the first was the inconsistency of the internal control (IC) when added to the PCR mixture. This point was not found under routine conditions when the IC was added during the extraction step. The second is a lack of practicality, as the mixture is distributed over several vials, requiring numerous pipetting operations. Overall, this study provides useful information for the molecular diagnosis of toxoplasmosis; the analytical and clinical performances of the Sacace PCR kit were satisfactory, the kit having sensitivity and specificity similar to those of expert center methods and being able to detect low parasite loads, at levels where multiplicative analysis gives inconsistently positive results. Finally, the study recommends multiplicative analysis in particular for amniotic fluids, aqueous humor, and other single specimens.